220 



SCIENCE. 



{Vol. XIV. No. 347 



they are shaken ; but it is almost — I might say, quite — impossi- 

 ble to believe that when this sifting has only been commenced, as 

 it would be in the short exposure to which a camera picture is sub- 

 mitted, the substance deposited to build up the image by purely 

 chemical means would be so obliging as to deposit in that the par- 

 ticular size of particle which should give to the image the color of 

 the nucleus on which it was depositing. I am aware that in the 

 early days of photography we heard a good deal about curious re- 

 sults that had been obtained in negatives, where red brick houses 

 were shown as red, and the blue sky as bluish. The cause of these 

 few coincidences is not hard to explain, and would be exactly the 

 same as when the red brick houses were shown as bluish, and the sky 

 as red, in a negative. The records of the production of the latter 

 negatives are naturally not abundant, since they would not attract 

 much attention. , I may repeat, then, that photography in natural 

 colors by a printing-out process — by which I mean by the action 

 of light alone — is not only possible, but has been done, but that 

 the production of a negative in natural colors from which prints in 

 natural colors might be produced, appears, in the present state of 

 our knowledge, to be impossible. Supposing it were not impracti- 

 cable, it would be unsatisfactory, as the. light with which the pic- 

 ture was impressed would be very different from that in which it 

 would be viewed. Artists are fully aware of this difficulty in paint- 

 ing, and take their precautions against it. 



The nearest approach to success in producing colored pictures 

 by light alone is the method of taking three negatives of the same 

 subject through different-colored glasses, complementary to the 

 three color-sensations which together give to the eye the sensations 

 of white light. The method is open to objection on account of the 

 impure color of the glasses used. If a device could be adopted 

 whereby only those three parts of the spectrum could be severally 

 used which form the color-sensations, the method would be more 

 perfect than it is at present. Even then, perfection could not be 

 attained, owing to a defect which is inherent in photography, and 

 which cannot be eliminated. This defect is the imperfect repre- 

 sentation of gradation of tone. For instance : if we have a strip 

 graduated from what we call black to white (it must be recollected 

 that no tone can scientifically be called black, and none white), and 

 photograph it, we shall find that in a print from the negative the 

 darkness which is supposed to represent a gray of equal mixtures 

 of black and white by no means does so unless the black is not as 

 black nor the white as white as the original. The cause of this un- 

 truthfulness in photography has occupied my attention for several 

 years, and it has been my endeavor to find out some law which 

 will give us the density of a silver deposit on a negative correspond- 

 ing with the intensity of the light acting. I am glad to say that at 

 the beginning of this year a law disclosed itself, and I find that the 

 transparency of a silver deposit caused by development can be put 

 into the form of the law of error. 



This law can be scarcely empiric, though at first sight it appears 

 that the manipulations in photography are so loose that it should 

 be so. It is this very looseness, however, which shows that the 

 law is applicable, since in all cases I have tried it is obeyed. That 

 there are theoretical difficulties cannot be denied, but it is believed 

 that strictly theoretical reasoning will eventually reconcile theory 

 with observation. 



This want of truth in photography in rendering gradation, then, 

 puts it out of the range of possibility that photography in natural 

 colors can ever be exact, or that the three-negatives system can 

 ever get over the difficulty. 



One of the reproaches that in early days was cast at photography 

 was its inability to render color in its proper monochromatic lumi- 

 nosity. Thus, while a dark blue was rendered as white in a print 

 (that is, gave a dense deposit in a negative), bright yellow was ren- 

 dered as black in a print, or nearly so (that is, as transparent or 

 nearly transparent glass in the negative). To the eye the yellow 

 might be far more luminous than the blue, but the luminosity was 

 in the photograph reversed. I need scarcely say that the reason of 

 this want of truth in the photograph is due to the want of sensi- 

 tiveness of the ordinarily used silver salts to the least refrangible 

 end of the spectrum. Some fifteen years ago Dr. H. W. Vogel an- 

 nounced the fact that when silver salts were stained with certain 

 dyes they became sensitive to the color of the spectrum, which the 



dyes absorbed. This at once opened up possibilities, which, how- 

 ever, were not at once realized, owing perhaps to the length of ex- 

 posure required when the collodion process was employed. Shortly 

 after the gelatine process was perfected, the same dyes were ap- 

 plied to plates prepared by this method, which, although they con- 

 tained the same silver salts as the old collodion process, yet per se 

 were very much more sensitive. A new era then dawned for what 

 has been termed " isochromatic " and " orthochromatic " photog- 

 raphy. The dyes principally used are those belonging to the 

 eosine group and cyanine ; not the ordinary cyanine dye of com- 

 merce, but that discovered by Greville Williams. For a dye to be 

 of use in this manner, it may be taken as an axiom — first pro- 

 pounded by the speaker, it is believed — that it must be fugitive, or 

 that it must be capable of forming a silver compound. The more 

 stable a dye is, the less effective it is. If we take as an example 

 cyanine, we find that it absorbs in the orange and slightly in the 

 red. If paper or collodion stained with this coloring-matter be 

 exposed to the action of the spectrum, it will be found that the dye 

 bleaches in exactly the same part of the spectrum as that in which 

 it absorbs, following, indeed, the universal law I have already alluded 

 to. If a film containing a silver salt be dyed with the same, it will 

 be found, that, while the spectrum acts on it in the usual manner, — 

 viz., darkening it in the blue, violet, and ultra-violet, — the color is 

 discharged where the dye absorbs, showing that in one part of the 

 spectrum it is the silver salt which is sensitive, and that in the 

 other it is the coloring- matter. If such a plate, after exposure to 

 the spectrum, be developed, it will be found that at both parts a 

 deposit of silver takes place ; and, further, when the experiment is 

 carefully conducted, if a plate with merely cyanine-colored collodion 

 be exposed to the spectrum and bleached in the orange, and after 

 removal to the dark-room another film containing a silver salt be 

 applied, and then a developer, a deposit of silver will take place 

 where the bleaching has occurred. This points to the fact that the 

 molecules of a fugitive dye, when altered by light, are unsatisfied, 

 and are ready to take up an atom or atoms of silver ; and other 

 molecules of silver will deposit on such nuclei by an action which 

 has various names in physical science, but which I do not care to 

 mention. This is the theory which I have always advocated ; viz., 

 that the dye by its reduction acts as a nucleus on which a deposit 

 of silver can take place. It met with opposition ; a rival theory 

 which makes the dye an " optical sensitizer " — an expression 

 which is capable of a meaning which I conceive contrary to physi- 

 cal laws — being run against it. The objection to what I may call 

 the nucleus theory is less vigorous than it has been, and its diminu- 

 tion is due, perhaps, to the more perfect understanding of the mean- 

 ing of each other by those ^engaged in the controversy. To my 

 mind, the action of light on fugitive dyes is one of the most inter- 

 esting in the whole realm of photography, as eventually it must 

 teach us something as to the structure of molecules, and add to the 

 methods by which their coarseness may be ascertained. Be the 

 theory what it may, however, a definite result has been attained, 

 and it is now possible to obtain a fair representation of the lumi- 

 nosity of colors by means of dyed films. At present the employ- 

 ment of colored screens in front of the lens, or on the lens itself, is 

 almost an essential in the method, when daylight is employed ; but 

 not till some dye is discovered which shall make a film equally 

 sensitive for the same luminosity to the whole visible spectrum 

 will it be possible to make orthochromatic photography as perfect 

 as it can be made. The very fact that no photograph of even a 

 black and white gradation will render the latter correctly, must of 

 necessity render any process imperfect, and hence in the above 

 sentence I have used the expression "as perfect as it can be made." 

 The delineation of the spectrum is one of the chief scientific ap- 

 plications to which photography has been put. From very early 

 days the violet and ultra-violet end of the spectrum have been 

 favorite objects for the photographic plate. To secure the yellow 

 and red of the spectrum was, however, till of late years, a matter 

 of apparently insurmountable difficulty ; while a knowledge of that 

 part of the spectrum which lies below the red was only to be 

 gained by its heating effect. The introduction of the gelatine pro- 

 cess enabled the green portion of the spectrum to impress itself on 

 the sensitive surface ; while the addition of various dyes, as before 

 mentioned, allowed the yellow, the orange, and a portion of the red 



