128 



NA TURE 



[December ii, 1902 



As a supplement to Mr. Dresser's account of the work 

 he has carried on, we have added a summary of the 

 account of the process given by Sir Henry Trueman Wood 

 at the Royal Society's conversazione last May, when the 

 rationale of the process was demonstrated. 



The Three-Colour Photographic Process. 



To produce a photograph in colour direct from nature 

 has for many years past been the dream and cherished 

 aim of many photographers, but, so far as I can ascertain, 

 these efforts have not met with success. By a happy 

 combination, however, of the camera and the printing 

 press, the so-called three-colour process has been so far 

 perfected as to have become a commercial success, and, 

 though still, perhaps, in its infancy, bids fair to become a 

 serious rival to chromolithograph)', not only on account 

 of its accuracy, but also because of its cheapness. 

 Moreover, in the case of a larger number of copies being 

 required, the total cost is considerably below that of 

 chromolithography. 



Upwards of twenty years ago, when the publication of 

 my "Birds of Europe" was drawing to a close, I com- 

 menced to collect materials for a companion work on 

 the eggs of European birds. When, however, it 

 arrived at a question of illustrations, I found that I could 

 not get plates sufficiently well and cheaply executed by 

 any then known process. Besides which I could find 

 no artist who could reproduce eggs in water-colour 

 satisfactorily, and indeed, at the present time, I 

 know of only one, a Danish artist, who can paint eggs 

 with sufficient accuracy, and he is at present engaged 

 on the illustrations for the British Museum "Catalogue 

 of Eggs." Nor can he copy all sorts of eggs correctly, 

 for in some species the markings are so minute and 

 varied that no artist could exactly reproduce them. 



In 1900, however, I saw a plate of fruit, photographed 

 directly from the object, without the intervention of an 

 artist, and reproduced by the three-colour process, 

 which gave me the idea that it would be specially 

 suited for the reproduction of natural history objects, 

 and I at once commenced a series of experiments 

 to test it with the assistance of Mr. I. D. Geddes, 

 manager to Messrs. Andre and Sleigh, Ltd., of Bushey, 

 Herts, and to his active cooperation I am indebted for 

 the success that has crowned my endeavours. To pro- 

 duce the coloured picture three negatives are made from 

 the objects on specially sensitised plates, which are 

 exposed through "light filters" placed behind the lens. 

 These filters separate out the colours of the objects into 

 what are known as the primary colours — approximately 

 red, blue and green. The negatives so obtained are then 

 employed in the usual manner for the production of 

 half-tone blocks — that is to say, each of the three pictures 

 representing the separated red, blue and green images 

 are etched as type blocks on copper for printing in the 

 ordinary press, and it must be noted that the pictures as 

 engraved on the copper blocks are made up of very fine 

 dots. The plates are printed in the colour complementary 

 to that of the filter through which each was taken, ie. 

 the red-filter picture in blue, the green in red and the 

 blue in yellow. The printing of the plates is effected on 

 three presses, one for each colour ; the yellow image is 

 first printed, then the red over the yellow printing, and, 

 lastly, the blue over the red and yellow, and in each case 

 the colour is allowed to dry before the next colour is 

 printed. The registration of one colour over the other 

 must be accurate, otherwise a blurring of the whole pic- 

 ture occurs. The colours used for printing are mixed 

 each to a standard tint, which is only departed from in 

 very exceptional cases. 



The length of exposure for the process varies very 

 much according to the conditions. As carried out for me 

 by Messrs. Andre and Sleigh, in which the pictures were 



NO. 1728, VOL. 67] 



taken with a light-filter, a prism and a ruled screen inter- 

 posed, the exposures were very long, the blue, approxi- 

 mately, ten to fifteen minutes, the green thirty to 

 forty minutes, and the red nearly two hours. This 

 process is eminently adapted for the copying of paint- 

 ings, but the sole aim of the experiments made has been 

 with a view to reproduce natural history objects, and 

 more especially eggs, without the intervention of an 

 artist. 



Mammals cannot be photographed from living exam- 

 ples, as the exposure required is too long, and can only 

 be done from paintings, for the reproductions are so 

 very accurate that if photographed from stuffed speci- 

 mens it is painfully apparent that they were stuffed. 

 The same may be said with regard to birds, but when 

 photographed from well-stuffed skins every character is 

 most accurately reproduced, and such plates are con- 

 sequently of extreme scientific value. Some fishes and 

 crustaceans retain their colours for some time after 

 death, whereas others fade almost immediately ; the 

 former of these can in most cases be reproduced from 

 the specimens direct, but as regards the latter it will 

 be necessary to employ an artist. 



Shells of all kinds are specially adapted for this pro- 

 cess, as colour-photography brings out even the 

 bright iridescent colourings so characteristic of some 

 species. 



Flowers and plants, however, present serious difficul- 

 ties, owing also to the long exposure required. Cut 

 flowers will move and fade, and growing plants are sure 

 also to move within three hours and thus spoil the pic- 

 tures. Butterflies, moths and other insects can be photo- 

 graphed from the specimens direct if these are perfect, 

 but they are often slightly damaged in catching, or in 

 drying they become somewhat distorted, and any slight 

 imperfection cannot be hidden, but is most faithfully re- 

 produced ; hence it is generally advisable to photograph 

 from water-colour drawings of these objects. 



Birds' eggs have chiefly occupied my attention, 

 and with these I have been most successful, so much 

 so that I purpose now to bring out my work on 

 eggs, illustrated by this process from the eggs direct, 

 without the intervention of an artist. At first I found a 

 difficulty with the shadows, and tried the effect of a dark 

 background ; but as this took from the characteristic 

 colours of some species, I had to revert to a pale back- 

 ground, and by degrees have overcome the difficulties, 

 as will be seen from the plate accompanying the 

 present article. The eggs figured on this plate are as 

 follows : — 



Figs. 1, 2, 3, eggs of the Lesser Kestrel, Falco cenchris j 

 Figs. 4, 5, eggs of the Honey Buzzard, Pernis apivorus ; 

 Fig. 6, egg of the Levant Sparrowhawk, Asti/r brevipesj 

 Fig. 7, egg of the Shikra Sparrowhawk, Astur bmiuis ,- 

 Figs. 8, 9, 10, eggs of the Blackwinged Kite, El, nan 

 coemleus. All these specimens have been selected to 

 show the greatest variation in these eggs, and also to 

 test the process. H. E. Dresser. 



Principles of Three-Colour Photography. 1 



The reproduction of the camera picture in its natural 

 colours is still an unsolved problem, for Lippmann's 

 results can hardly be said to have passed the experi- 

 mental stage. They still lack practical application. 

 All that can be done by photographic means is to select 

 and combine colours, so as to produce an approximately 

 correct reproduction of the colours of any natural object. 

 The colour itself must be provided by the use of dyes, 

 stains or pigments. 



The principal application of the three-colour process 



\ Subject-matter of a demonstration given at llie conversazione of the 

 Royal S ciety on May 14 by Sir H. Trueman Wood. 



