The numerical expression of plate sensitivity has become in many cases a means of commercial advertisement, numbers being given to emul sions without previous consultation with a firm's technicians, so that only very rough values can be attached to indications which are given." An agreement between German makers of sensitive emulsions defined, in 1932, under the name of DIN 4512 (Deutsche Industrie Normen), a method of indicating the speed of the usual negative emulsions, especially films for amateurs. A sample is exposed for 1/20 second behind a. neutral grey scale of which the densities pro gress by o-I. This scale receives an illumina tion of 40 lux (artificial sunlight). After very prolonged development (so as to obtain the number most favourable to the emulsion) in a specified developer, the speed of the emulsion is stated by the density of the sensitometric screen under which there has been obtained a density superior by 0.1 to the fog density, without this latter exceeding 0-4. For instance, if the desired density was obtained (average of four tests) under patch No. i8 of the screen (density 1.8), the speed is stated as I8c7 Eo DIN (with a tolerance of 3°,110, corresponding to a variation from single to double). There is, unfortunately, no relation between the speeds of different emulsions, developed in conditions such as to give a usable negative, and the speeds assigned to them after the excessive development (which exaggerates contrast and graininess) prescribed for the application of the DIN method.
The makers of different exposure-meters (324 326), Watkins, Wynne,. etc., publish from time to time, for use with these instruments, tables showing for the best-known types of emulsions coefficients of sensitivity which represent gener ally quite satisfactorily the average properties of each type.
It may be mentioned that emulsions manu factured successively under the same name by the same maker may have markedly different characteristics, the differences being sometimes as great as those between emulsions of different makes.
Except for the photography of subjects in very rapid motion or animated subjects in bad light, a quality of the emulsion far more impor tant than its speed is its latitude of exposure, the relation between extremes of exposure for the correct reproduction of a subject of medium contrast. The double coating of a support with a very rapid emulsion upon a slow one so that the latter comes into action when the upper one reaches solarization often permits, in the case of a subject of which the extreme luminosities are in the relation of about I : 30, variations in exposure greatly exceeding a range of Ioo to I. The characteristic curve of these double coated emulsions is usually broken.
204. Reversal and Solarization. When a series of increasing exposures, extending over a very wide range, are given to an emulsion, it is found after development that the resulting density first increases, reaches a maximum, and then diminishes until it becomes almost nothing. In 188o Janssen suggested the existence of a second range of increasing densities beyond the point where the blackening had decreased to nothing. Thus, in photographing a natural object, it is possible, according to the time of exposure given, to obtain either a normal nega tive, or an almost uniform high density (first neutral zone), or a more or less satisfactory positive (image reversed by solarization), or, lastly, an almost uniform and very weak density (second neutral zone). Generally speaking, the
time of exposure necessary to obtain a direct positive on an ordinary emulsion is Loc.° times that necessary to give a satisfactory but this process is uncertain and is used only with special emulsions (§ 439).
The reversal of images on gelatino-bromide emulsions may be avoided, or at least reduced, by impregnating the emulsion with different substances (sodium nitrite, salts of hydrazine, or paraphenylene-diamine, etc.), which absorb the bromine liberated from the silver bromide by the prolonged action of the light ; it is made easier by partial iodization of the emulsion (A. Charriou and S. Valette, 1935).
Further, the effects of solarization may be destroyed by treating the emulsion, before chemical development, with solutions of chromic acid, or by physical development after fixation.
Abnormal reversals occur in certain special cases. A very short exposure (of the order of Ii5o,000th of a second) to a very intense light lowers the sensitivity of the plate and prevents it from fogging when again exposed to the light 2 (Clayden effect).
Reversal of the image is also sometimes noticed in developing a normally-exposed plate by means of a light which is not quite safe ; the light prints a positive of the normally developing negative on the lower layers of the emulsion, and this positive image may become preponderant. A similar phenomenon is observed when white light is momentarily admitted to the room while the image is still weak (Sabatier effect), due to the desensitization of the emulsion by the oxidation products of the developer (§ 330). This reversal occurs also if the developer is an alkaline solution of hydrogen peroxide (G. W. W. Stevens and R. G. Norrish, 1934). The result is the same whichever face of the sensitive layer is exposed to light for the second exposure (A. P. H. Trivelli, 205. The Accuracy of Photographic Images. Each brightly illuminated point of a sensitive film acts towards the surrounding grains as a secondary source of light and thus produces a developable image at points beyond the part directly illuminated until a point is reached where, due to absorption, the quantity of light transmitted is less than the threshold value. By the application of the laws of absorption, :Eery (1899) showed, and experiments have verified, that the enlargement of the image follows an arithmetical progression when the quantities of light increase in geometrical progression. This phenomenon is known as irradiation. However, a plate must be over-exposed at least ten times to enlarge to double its width a line 1/125th of a millimetre (this being really a spreading of 1/250th of a millimetre on each side of the line). This increase of thickness has really important consequences only in scientific work of high accuracy and can be avoided by lightly staining the emulsion yellow or red and compensating for the absorption of the light by the dye by increasing the time of exposure.