92 



NATURE 



[November 28, 1895 



the red and violet curves of Maxwell being based on different 

 wave-lengths to those used by Kcenig, the fact of grave in- 

 accuracy is certain. This fact will appear if the spectrum is 

 photographed according to Maxwell's curves. The blue-green 

 will then be found to be reproduced too yellow in tone. 



In order to apply the colour sensation curves of Koenig to the 

 photographic method, we have to find by trial examinations of 

 his curves the green most suitable for backing the "green" 

 positive ; for we see that the several green wave-lengths excite 

 very different amounts of red and violet sensation. We find as 

 suitable a wave-length a little to left side of the E line, about 

 550 puL. If we take this colour to back the green positive, 

 we must, in order to find the correct red and violet curves 

 which are to control the densities of the red and violet images, 

 replot the red and violet curves with allowance for the pro- 

 portionate amounts of red and violet which will be carried 

 to all those points where in the image of the spectrum the 

 green curve operates. The red and violet curves must be 

 lowered by amounts obtained by ascertaining from the height of 

 the green curve at any point the amount of red and violet sensa- 

 tions excited by the amount of our selected green present at that 

 point. The final curves are shown in the slightly altered violet 

 sensation curve, the original green sensation curve, and the con- 

 siderably lowered red-mixture curve (as it may be called) ; the 

 full line in all cases representing the applicable curves. It is 

 seen that the amount of negative colour (which cannot be 

 realised) is small. Although it is possible that the compound 

 nature of our green sensations will deny absolute accuracy to this 

 method of colour photography, still my own results on the 

 curves just described, and the results of Ives and others on 

 modified Maxwell curves, appear to show that a degree Of 



accuracy baffling the criticism of the ordinary untrained eye 

 may be attained, and that in the reproduction of the most 

 complex tints. 



The symmetry of the derived curves renders their application 

 easy. The transmission of light through a pigment is not 

 limited generally to a small group of predominant wave-lengths, 

 but falls off uniformly at either side in the directions of longer 

 and shorter waves. If we choose the pigments used on the 

 analysing screens so that their predominant transmissions are at 

 three points in the spectrum indicated by the axis of symmetry 

 of the three curves, these being nearly symmetrical, very accurate 

 results are obtained. The positions of these axes of symmetry 

 are shown by the vertical dotted lines. Accordingly, I make the 

 colour of the red-taking screen that of the spectrum at a point 

 displaced to the red side of D by about one-sixth the interval D 

 to C ; for the green and violet-taking screens the correct tints 

 are found in the same manner by scaling from the figure, 

 (iood results are thus obtained, but I do not assert that these 

 details of procedure are final. 



Any method of photography in natural colours must possess 

 the characteristics not only of accuracy of colour rendering, but 

 also of convenience of application and permanency of colour, 

 if it is to possess value as a scientific method. For use under 

 the various circumstances of travel the naturalist requires a 

 method no more cumbersome than the present dry plate. In 

 the method of composite colour photography, as described, the 

 ordinary camera will not serve. The cumbersome necessity of 

 obtaining three images remains, and subsequently no concrete 



NO. I 36 I, VOL. 53] 



image in natural colours is actually obtained. One can only be 

 realised by triple projection upon a screen, or by using some 

 optical contrivance which, by the aid of reflectors, enables all 

 three images to be simultaneously projected upon the retina. 



I now proceed to describe a mode of applying the foregoing 

 principles which is free of the objection of cumbersomeness, 

 and which enables us to realise a concrete image in transparent 

 colours. A plate is finally produced which may be held in the 

 hand, regarded against the light, and which bears an image of the 

 object in natural colours, or such as are so nearly accurate as to 

 seem so to the eye. In this new method there is but the one 

 image photographed. The ordinary camera, lens, and backs, 

 &c., are used without modification. The first-class isochromatic 

 plates in the market, which are sensitised down to the C red, 

 will give very good results. 



In the new method the idea is to carry the application of 

 physiological principles still further, and divide up the plate like 

 a hypothetical subdivision of the retina, so that ail over the plate 

 there should be minute regions uniformly distributed wherein the 

 sensitive silver salt is excited to become reduced to the " photo- 

 genic " material in the same degree in which the sensations of 

 redness, greenness, violetness, would have been actually excited 

 in the several nerves of the retina had the image been formed 

 upon it. Development builds upon this photogenic material the 

 denser silver deposit, and ultimately in the positive the amounts 

 of the sensations are registered in the degrees of transparency of 

 the successive regions. The lined screen which can bring about 

 this I can show you in the microscope. It consists of closely 

 ruled adjacent lines in orange, green and violet tints. This screen, 

 applied closely to the sensitive .surface, analyses the image in the 

 camera. The screens I have used hitherto are coarse, about 200 

 lines to the inch, and even with this coarseness will show plainly, I 

 regret to say, the imperfections of the only apparatus at my com- 

 mand in preparing these screens. I may observe, in passing, that 

 the colours are ruled on in pigments made up as inks in gelatine 

 and gum arable or dextrine, and upon plates coated with a pre- 

 liminary layer of gelatine. Such lines may be put on so close 

 as 800 or 1000 to the inch. With between 300 and 400 to the 

 inch, however, the eye is no longer annoyed by the structure of 

 the plates. The lines may also be ruled on celluloid or on 

 translucent paper. 



The appearance of both negative and positive obtained is 

 interesting. One would hardly at first sight distinguish be- 

 tween them and the ordinary images. But a lens readily 

 shows the difference. Recalling now that the lines upon the 

 positive register in their degrees of transparency the degrees in 

 which the three-colour sensations would have been excited, it 

 becomes apparent that to complete the physiological parallel we 

 must convert these degrees of transparency to quantities of the 

 red, blue and violet colour sensations. This is done by a second 

 screen, which carries red, green and violet lines to the same 

 gauge as the taking screen. We apply this to the positive, and 

 as we move it over the image, waves of every tint of colour 

 appear till that position is reached where the red lines fall over 

 the lined areas recording red sensation, and so for the others. 

 The picture now suddenly appears in vivid colour and with all 

 the realism and relief conferred by colour perspective. 



A picture of wallflowers taken through a dichromatic screen, 

 the red and green sensations only being photographed, is of 

 interest as realising the appearance of the object to a violet-blind 

 eye. The rich reds and browns appear unaffected ; the greens 

 are, however, somewhat unnatural. A photograph of the spec- 

 trum shows the range of colour from the C red to the H lines. 

 The blue-green is, however, defective. It was taken according 

 to Maxwell's curves. Photographs of burnished metallic objects, 

 as a brightly lacquered microscope, reproduce the metallic lustre ; 

 and one of an uranium glass bowl, reproduces the characteristic 

 dichroism and fluorescent appearance of the glass when seen by 

 daylight. That every shade of colour can be reproduced, how- 

 ever complex, is shown by two portraits, one from life, and one 

 a copy of a water-colour drawing boasting very esthetic shades 

 of brown and olive. A great variety of bright sunlit colours 

 appear in a view at the Trinity College athletic sports, wherein 

 the scarlet uniforms of the military band, the green of the grass, 

 and the blue sky, recall the vivid appearance of the image on 

 the ground glass screen of the camera. The colour perspective 

 in such pictures adds greatly to the reality and relief. The 

 faithful reproduction of texture, as in the case of some pansies, 

 where the velvety browns and purples of the originals reappear, 

 or as in the case of the wallflowers, reminds us how much is 



