April 30, 1896] 



NATURE 



617 



COLOUR PHOTOGRAPHY. 



^PJ IK scientific event of last week was the description and 

 demonstration of colour photography given by Prof. 

 Lippmann before the Royal Society. On the occasion of the 

 centenary celebrations of the Institute of France last year. Lord 

 Kelvin invited Prof. Lippmann to give the Royal Society 

 an account of his researches on photography in colours, and last 

 Thursday's meeting was the result. The methods employed by 

 Prof. Lippmann are well known among men of science, but few 

 of the Royal Society were prepared to see such remarkable 

 results as those obtained and exhibited by the distinguished 

 French physicist. The honour and fine feeling which such visits 

 bring to the Society, and the extreme interest aroused, should 

 help to make similar occasions of more frequent occurrence. We 

 print Prof. Lippmann's lecture below, and our only regret is 

 that it cannot give an at all adequate conception of the striking 

 achievement with which it deals. 



The problem of colour photography is as old as photography 

 itself. The desire of fixing the colours as well as the design 

 of the beautiful image thrown on the screen of the camera, very 

 naturally occurred to the earliest observers. Since the begin- 

 ning of this century three distinct solutions of the problem have 

 been realised. 



The first solution, not quite a complete one, is founded on the 

 peculiar properties of a silver compound, the violet subchloride 

 of silver. E. Becquerel (i860) converted the surface of a 

 daguerreotype plate into this silver compound, and by projecting 

 on it the image of the solar spectrum, and other objects, obtained 

 good coloured impressions. Poitevin substituted paper for the 

 silver plate as a substratum. No other substance has been 

 discovered that can play the part of the subchloride of silver. 

 Moreover the image is not fixed, in the photographic sense of 

 the word ; that is, the coloured impression is retained for any 

 length of time in the dark, but it is blotted out by the action of 

 daylight. The reason of it is this : the Becquerel images are 

 formed by coloured silver-compounds, which remain sensible to 

 light ; so that they are destroyed by the continued action of light, 

 in virtue of the same action which gave them birth. Despite 

 the numerous experiments made by Becquerel, Poitevin, Zenther, 

 and others, no suVjstance has been fouind that is capable of 

 destroying the sensibility of the subchloride for light without at 

 the same time destroying its colour. 



The second method for colour photography is an indirect one, 

 and may be called the three-colour method. It was invented 

 in France by Ch. Cros, and at the same time by M. Ducos du 

 Hauron (1869). German authorities claim the priority of the 

 idea for Baron Bonstetten. Three separate negatives (colour- 

 less) are taken of an object through three coloured screens. 

 From these three positives (equally colourless) are made ; and, 

 lastly, the colour is supplied to these positives by means of 

 aniline dyes or coloured »inks. Thus three coloured mono- 

 chromatic positives are obtained, which by superposition give 

 a coloured image of the model. In the ingenious process lately 

 invented by Prof. Joly, the three negatives, and apparently the 

 corresponding three positives, are obtained interwoven on one 

 and the same plate. ' The three-coloured method can give a 

 very good approximation to the truth, and has probably a great 

 future before it. We may call it, nevertheless, an indirect 

 niethod, since the colours are not generated by the action of 

 light, but are later supplied by the application of aniline dyes or 

 other pigments. Moreover, the choice of these pigments, as 

 well as of the coloured screens through which the negatives 

 have been obtained, is in some degree an arbitrary choice. 



The third and latest method by which colour photography has 

 been realised is the interferential method, which I published in 

 1S91, and the results of which I beg to lay before you this 

 evening. It gives fixed .images, the colours of which are due 

 to the direct action of the luminous rays. 



For obtaining coloured photographs by this method, only two 

 conditions are to be fulfilled. We want (i) a transparent grain- 

 less photographic film of any kind, capable of giving a colour- 

 less fixed image by the usual means ; and (2) we want a 

 metallic mirror, placed in immediate contact with the film 

 during the time of exposition. 



A mirror is easily formed by means of mercury. The photo- 

 graphic plate being first enclosed in a camera-slide, a quantity 

 of mercury is allowed to flow in behind the plate from this small 

 reservoir, which is connected with the slide by a piece of india- 



NO. 1383, VOL. 53] 



rubber tubing. 1 The slide is then adapted to the camera, and 

 the action of light allowed to take place. After exposure the 

 slide is separated from the camera, the mercury reservoir lowered 

 so as to allow the mercury to flow back into it ; the photo- 

 graphic plate is then taken out, developed and fixed. When 

 dry, and examined by reflected light, it appears brilliantly 

 coloured. 



The sensitive film may be made either of chloride, iodide, or 

 bromide of silver, contained in a substratum either of albumen, 

 collodion, or gelatine. The corresponding developers, either 

 acid or alkaline, have to be applied ; the fixation may be cyanide 

 or bromide of potassium. All these processes I have tried with 

 success. For instance, the photograph of the electric spectrum 

 now projected before your eyes, has been made on a layer of 

 gelatino-bromide of silver, developed with amidol, and fixed with 

 cyanide of potassium. 



As you see, bright colour photographs may be obtained with- 

 out changing the technique of ordinary photography ; the same 

 films, developers and fixators have to be employed ; even the 

 secondary operations of intensification and of isochromatisation 

 are made use of with full success. The presence of the mirror 

 behind the film during exposure makes the whole difference. 

 From a chemical point of view nothing is changed, the result 

 being a deposit of reduced silver left in the film, a brownish, 

 colourless deposit. And yet the presence of a mirror during 

 exposure causes the colourless deposit to show bright colours. 

 Of course we want to know how this is done ; we require to^ 

 understand the theory of those colours. 



We all know that colourless soap-water gives brilliant soap- 

 bubbles ; the iridescence of mother-of-pearl takes birth in 

 colourless carbonate of lime ; the gorgeous hues of tropical 

 birds are simply reflected from the brownish substance which 

 forms the feathers. Newton discovered the theory of these 

 phenomena, and subjected them to measurement ; he invented 

 for the purpose the experiment called by the name of Newton's 

 rings. Newton showed, as you know, that when two parallel 

 reflecting surfaces are separated by a very short interval, and 

 illumined by white light, they reflect only one of the coloured 

 rays which are the constituents of white light. If, for instance, 

 the interval between the reflecting surface is only nr5T!D of a 

 millimetre, violet rays are alone reflected, the rest being 

 destroyed by interference ; that is, the two surfaces send back 

 two reflected rays whose vibrations interfere with one another, 

 so as to destroy every vibration except that which constitutes 

 violet light. If the interval between the reflecting surfaces be 

 augmented to tu^tt^ of a millimetre, the destruction of vibration 

 takes place for every vibration except that of red light, which 

 alone remains visible in this case. 



If we consider now this photograph of the spectrum, and 

 especially the violet end of the image, we find that this is formed 

 by a deposit of brown reduced silver. In the case of an 

 ordinary photograph, this deposit would simply be a formless 

 cloud of metallic particles ; here the cloud has a definite, strati- 

 fied form ; it is divided into a number of thin, equidistant strata, 

 parallel to the surface of the plate, and xTi%-G-Q of a millimetre 

 apart. These act as the reflecting surfaces considered by New- 

 ton, and as they are at the proper distances for reflecting violet 

 rays, and these alone, they do reflect violet rays. 



The red extremity of the photograph is equally built up of 

 strata which act in a like manner ; only their distance intervals 

 here amount to Tsf jnr of a millimetre, and that in the proper 

 interval for reflecting red light. The intermediate parts of the 

 spectral image are built up with intermediate values of the 

 interval, and reflect the intermediate parts of the spectrum. 



The appearance of colour is therefore due to the regular 

 structure above described, imprinted on the photographic 

 deposit. The next question is — How has this very fine, peculiar, 

 and adequate structure been produced ? 



It is well known that a ray of light may be considered as a 

 regular train of waves propagated through the ether, in the same 

 way as waves on the surface of water. The distance between 

 two following waves is constant, and termed the wave-length ; 

 each sort of radiation, each colour of the spectrum, being 

 characterised by a particular value of the wave-length. Now, 

 when a ray of light falls on a sensitive film, this train of waves 

 simply rushes through the film with a velocity of about 300,000 

 kilometres per second ; it impresses the film more or less 

 strongly, but leaves no record of its wave-length, of its particular 



1 The glass of the photographic plate has to be turned towards the ob- 

 jective, the film in contact with the metallic mirror. 



