HOWARD FARMER discovered that if a silver image imbedded in gelatine was immersed in a solution of a bichromate a reaction took place and the bichromate was reduced, so that the gelatine surrounding each minute particle of silver became insoluble in hot water. On this has been based a process for obtaining a gelatine relief, which can be stained up and used as the matrix for the transfer of dyes to gelatined paper. This is obviously the same principle as that involved in the carbon process, with the added advantage of being independent of daylight. Obviously it suffers from exactly the same disadvantage, that is, if we expose from the front in the usual way, the delicate details and half tones will be unsupported and will wash away in the development. We must, therefore, print through the back of the support. In the case of celluloid this is, of course, of no moment, as the thickness of the celluloid is not sufficient to make the image markedly diffuse; but if it is desired to use glass plates, the thickness of the glass causes want of sharpness in the print. This may not be sufficient to be objectionable to some people, but unless the glass of all three plates is of exactly the same thickness, the diffusion of the images will differ and the individual pulls be unequally sharp. This trouble in using glass plates may be gotten over by making the positives in the camera by copying, as is done when making lantern slides, only the transparency plate must be placed with its glass towards the lens and allowance for its thickness made in focusing the image.

In the case of celluloid films, while the same procedure may be adopted, one can print by contact, using an electric lamp. An opaque card with an aperture about one inch in diameter cut in it may be placed over the bulb, with a ground glass covering the hole, so as to obtain diffused light. The lamp should be placed four or five feet from the printing frame. The exposure under these conditions is naturally longer than in contact printing, but it is not unduly prolonged.

Any slow transparency plate may be used, but contrast or photomechanical plates should be avoided, as they readily give too much contrast. Any developer except pyrogallol may be used, and this should be avoided on account of its tanning action and staining. The images should not be too dense, otherwise the pull from the matrix may be wanting in details in the highlights. Naturally the positive must also be free from fog, for as the gelatine is rendered insoluble wherever there is silver, and fog is nothing but silver generally distributed, it might happen that the fog is sufficiently bad to give insoluble gelatine and consequently color where it is not wanted.

The positive should preferably be fixed and washed as usual, though this is not absolutely essential, and then immersed in one of the following baths:

Chromic acid 5 g.

Potassium bromide 20 g.

Water 1000 ccm or:

Potassium bichromate 6 g.

Hydrochloric acid 10 ccm.

Water 1000 ccm.

These may be used over and over again, in fact until their action gets too slow; they should, however, be kept in the dark, when not in use.

The positive should be immersed in the solution in a dish, and the dish rocked until it is seen that it is bleached quite through to the back, on looking through the glass. Prolonging the immersion does no harm, but as a rule ten minutes is sufficient for any positive, no matter how dense it may be. The bleached positive is then immersed in water at about 400 C. (1050 F.), when the gelatine will be seen to gradually dissolve away, leaving the whitened silver image, which is either the bromide or chloride. Should the gelatine not dissolve easily, the temperature of the water may be raised two or three degrees; but it is not advisable to force development, as otherwise some of the isolated patches and the fine details may be washed away. The plate should then be rinsed in cold water to remove any adherent soluble gelatine, washed to remove any traces of the bichromate-acid bath, which would decompose the fixing bath, and immersed in a hypo bath until the white silver salt is dissolved. After this it requires but a brief washing, and can be dried or immediately stained up.

The author has a fancy for immediately staining up lightly, as this prevents any subsequent mistake as to which plate is which, as otherwise when dry the relief is so slight that it is difficult to distinguish between the various plates. The dyes used may be the same as those suggested for the pinatype process, the final transfer paper may be treated in the same way, and the method of superposition of the images also followed.

Exactly how many pulls may be taken from these matrices the author has never determined, but with reasonable care at least a couple of dozen may be made. This is probably as many as the average worker will require, but the life of the matrices may be prolonged by treating them with an alum solution; in this case, the primary development of the silver image must be pushed a little farther than usual. The alum bath, a ten per cent solution of ordinary alum or five per cent of chrome alum, should be applied to the reliefs after they are freed from hypo. They may then be washed for ten minutes and treated as suggested.

The advantage of this and similar processes over the pinatype process is that there is nothing but the gelatine image to stain up, so that no dye can be transferred to places where it should not be. Also, as all acid dyes readily stain the gelatine, and most of them transfer well, there is a much wider choice of dyes. Thus, one may use for the red picture, in addition to those already given, fuchsin, erythrosin, rhodamin G, various sorts of ponceau; for the yellow, acid yellow, naphthol yellow; and for the blue, diamine pure blue 3B, carmine blue and alizarine cyanol. In fact one can try out the various dyes sold for dyeing household materials, and these are very cheap.

If the final dye impression is weak, or wanting in details in the high-lights, the original positive was underexposed or underdeveloped, and conversely if the highlights show too much color, particularly in portraits, the positive was overexposed or overdeveloped. If the relative coloring is good, but too weak generally, the dyed matrix was not left long enough in contact with the paper. As with the pinatype process, corrections can be made by subsequent transfers from one or all of the plates.

It will be found advisable to dye up the matrix each time before transfer, and it is actually immaterial how long it is left in the dye solution, as the gelatine can only take up a certain amount of dye, and the only point to observe is that, the more dye it has absorbed, the more rapid the transfer to the gelatinized paper.