SOME SPECIAL PKOCESSES 279 



printing-frame some indication of the image is seen, and the 

 films must now be placed in water at a temperature of 90 to 

 105 F., until the unaffected portion of the image is dissolved out. 

 This may be carried out in the dark-room, but only an orange or 

 bright -yellow light is necessary. If the image does not appear 

 in relief in a few minutes, but still remains undissolved, the 

 temperature of the water may be raised until the desired effect 

 is obtained. When development is complete, the films are placed 

 in a weak solution of hyposulphite of soda say 1 in 40 until 

 all the bromide of silver not acted on is dissolved out, after which 

 they are washed. These two films now have to be stained with 

 the dyeing-solution supplied by the makers, and it is on the exact 

 degree to which this process is carried that success or failure 

 depends. Possibly the best method is to over-stain each of the 

 impressions, and then to immerse them in water and gradually 

 dissolve out the stain until sufficient remains. This method, 

 perhaps, allows of more precise control than an endeavour to 

 remove the films from the stain at the exact moment when they 

 have been treated for a sufficient length of time. A point to be 

 borne in mind is that unless the exposure is strictly correct for 

 each impression, there is a possibility of the images differing 

 somewhat in size, so that on binding them together it is 

 impossible to place them in exact register. 



After drying, the films are varnished, and have then to be 

 fastened together by a suitable cement, so that they exactly 

 overlie one another. The instructions on this point, as given 

 by the makers, are sufficiently clear. 



To obtain the greatest transparency, the films should be 

 cemented together with Canada balsam, and it is important 

 to ensure that no extraneous matter and no air-bubbles are 

 allowed to get in. The slide is then bound up, and if a mask 

 is desired it is preferably placed on the outside of the slide. 



If the whole process has been efficiently carried out, the 

 result is a reproduction in colour which is of perfect trans- 

 parency and colour-rendering. 



The other methods available are essentially different in 

 principle and practice. Instead of taking a separate negative 

 for each of the three primary colour records, the whole colour 

 effect is obtained at one operation and on one plate. Further, 

 no negative is taken but the direct colour positive is given by 



