Gradation of a Developed Photographic Image. 301 



opacities. These opacities may be measured and noted as "trans- 

 parencies," "opacities," or "densities," the last being the - log 



transparencies and the opacity . (These definitions of 



transparency 



opacity and density are those given by Hurter and Driffield, and are 

 generally understood as such in photographic literature.) Where 

 varying time exposures are given, it is convenient to start with some 

 unit of time, such as 10 seconds for the exposure of the first small 

 space on a plate, to double this exposure for the next small space, and 

 so on. When the measurements of transparency or density are made, 

 and the curve has to be plotted, the scale for the abscissa is conveni- 

 ently the number of the exposure — that is, the time of exposure in 

 powers of two. The ordinates are then set up as transparency of 

 deposit, total transparency being 100, or as densities which give the 

 absolute light cut off in terms of common logarithms. The curve 

 joining these different ordinates is in both cases approximately a 

 straight line for some distance, and, at each end, tends to become 

 parallel to the scale of abscissae, and this straight portion is taken as 

 representing the gradation of the plate. If the same plate be thus 

 exposed to different monochromatic lights, and the images developed 

 together and the density measured, it is easily seen from the plotted 

 curves if the "gradation " of the plate is the same in each case, since, 

 if they are, the straight portions of each curve should be parallel. 



[It may be noted that the less steep the gradation of a plate, the 

 greater will be the extremes of lights and shades in an object or view 

 that will be shown in a print, as the blackest tone obtainable on it 

 reflects about 3 per cent, of light. For this reason in sun-lighted 

 views, a plate showing a flat gradation should be employed, whilst in 

 those illuminated by a cloudy sky, a plate giving a steep gradation 

 should be used.] 



When obtaining the three negatives for three-colour printing where 

 the object is photographed through an orange, a bluish green, and a 

 blue screen, if there is much change in gradation caused by the 

 difference in the colour of the light reaching the plate, the true render- 

 ing of an object in its natural colours becomes an operation of extreme 

 difficulty. It was with a view to ascertain if some of the difficulties 

 which have been encountered in this process were due to difference in 

 gradation caused by the different coloured screens, that this research 

 was commenced some three years ago. Nearly two years ago, in an 

 article in ' Photography,' I indicated that a variation in gradation 

 due to difference in the monochromatic light in which the exposure 

 was made did exist, and some six months ago Mr. Chapman Jones, in 

 a paper communicated to the Royal Photographic Society, independ- 

 ently announced the same result from experiments made principally 

 with orthochromatic plates with light passing through various coloured 



Y 2 



