REVERSAL PROCESSES : METHODS FOR OBTAINING DIRECT POSITIVES 437. General Considerations. We will first men tion a method which cannot, properly, be con sidered as a reversal process, but which is some times employed for obtaining positive images directly. By developing so as to obtain silver, which is whitish by reflected light, and employ ing plates in which the emulsion is coated on a black or very dark support, the image appears as a positive. This method is chiefly used for making " ferrotypes," formerly known as tin types. An ordinary negative image on a transparent support may also be made to appear as a positive if the black reduced silver is con verted into a white salt of silver, and the plate is then given a black backing.
The methods of reversal, properly called, may be classified as follows s (a) By considerable over-exposure ; the image is solarized (§ 2o4) and develops directly as a positive. 4 (b) After normal development of a negative which has been normally exposed and has not been desensitized, only a part of the thickness of the emulsion will have been employed in obtaining the negative image ; the remainder is still sensitive to light (although its sensitivity may have been reduced) ; it is therefore possible, by exposing the negative to light, to produce in this residue a latent positive image which will remain exist, although weakened, after dis solving out the silver of the negative image, and may be developed so as to give the final positive image (C. Drouillard, (c) It has been noticed ( J. G. Capstaff, 1921) that, after dissolution of the first image, the speed of the remaining silver bromide varies consider ably from one point to another, being the less at each point the more the quantity of silver dissolved there. As a matter of fact, the various grains of silver each have different speeds, and in each spot it is the fastest grains which, dur ing the first exposure, are brought first into developable condition ; thus there remains in the image of the shadows the majority of the most rapid grains, while the slowest grains alone remain in the image of the high-lights. A uni
form exposure can therefore produce an effect practically identical to that of exposures, vari able from one point to another, corresponding to printing a positive under a negative ; what happens is almost as if the second exposure to light be made under the first (negative) image, although the latter has been removed. Develop ment and fixation are then proceeded with, just as with an image obtained under ordinary conditions.
(d) In a negative which has been developed but not fixed there exist two images. These are complementary to one another, but generally of very unequal quality ; one is the negative image, consisting of reduced silver, and the other is a positive image, consisting of the residual silver bromide. Whilst dissolving the silver bromide (fixation) only leaves the nega tive image, the removal of the silver by a solvent having no action on the bromide leaves the positive image of which it suffices to bring the silver bromide to the state of metallic silver or of some compound of suitable colour (C. Russell, 1862).
438. We shall deal only with methods (c) and (d) , the only ones in current use at present.
Reversal by a second, determined exposure (c) requires exact determination of the uniform exposure to be given (after dissolution of the first image) by preliminary tests on images taken under the same conditions as those to be treated. This method is only practicable in cinematography, in which case some pictures may always be taken from each scene. It is chiefly used for amateur films processed on con tinuous machines with a device for automatic adjustment of the exposing light according to the average amount of silver bromide remaining after the dissolution of the first Reversal by the " residue " method is better suited for treating individual images. Various means of augmenting its flexibility will be described.