There is occasionally some little misunder standing about the printing inks and the pro jection colours, but this can be cleared up by a very simple explanation. If a black-and white drawing is photographed, it is obvious that the whites will give the density in the negative, whilst the black, which is the colour in which it is wished to print, gives the shadows, or bare glass, so that obviously we print in the colour that does not act. In projecting a transparency from the negative we should project by white light ; therefore we should project by the colour that did act on the negative. Dealing thus with a coloured sub ject, we use a red screen that cuts out the blue, and therefore print from this negative in the colour that does not act on the negative, namely, blue ; using a green screen we cut out the red, and therefore print in red ; and with a blue screen, which cuts out the yellow, we print in yellow. For projection, as stated above, we project by the light by which the negatives were taken and, therefore, project the transparency taken through the red screen by red light, that taken by the green screen by green light, and that taken by blue by blue light.
Theoretically, there should be, of course, correspondence between the light cut out by the screens and the light reflected by the inks, but here there is generally a departure from theory, as the theoretically correct inks are not sufficiently permanent to light.
Three-colour work may be divided into photomechanical printing ; the superposition of dyed films or pigments, as in the carbon process, in which the principle of subtractive colour mixture is made use of ; and the optical synthesis methods, such as the chromoscope, three-colour projection, and screen-plate pro cesses. The production of filters for this work
is treated of under the heading of " Colour Screen or Filter." It is far better to use a panchromatic plate for all three exposures, as the gradations are then more likely to be the same, which may not be the case when plates sensitised for each particular section of the spectrum are employed. In no case should the negatives be harsh, but rather tending towards softness. With correct filters adjusted to the plate with which they are to be used and correct exposure, hand work on the negatives—except for the retouching of pure mechanical defects— should be avoided, as no one can tell from looking at a subject exactly how much of any one of the three printing colours is con tained in any colour of the original.
Pinatype is a three-colour process invented by Dr. Kdnig, in x9o5. Three negatives are first obtained through red, green and blue filters, and from these transparencies are made. Special gelatinised plates, sensitised with potassium bi chromate, are exposed under the positives and washed, being then soaked in blue, red and yellow dyes respectively, which are absorbed by the soft or image portions only. The three plates are pressed in turn in contact with moistened gelatinised paper, which takes up the dye ; a photograph, or strictly, print, in colours results.