Page:EB1911 - Volume 21.djvu/525

black wedges were used to make matches between the naked light and the same light after passing through the photographic opacity that had to be measured. In 1887, owing to the perfecting of the rotating sectors, which could be made to increase or diminish the apertures at pleasure during its rotation, the measurement of opacities became easy. The Rumford method of comparing the light through the deposit with the naked beam, using the sectors to equalize the illumination, was adopted, the deposit being placed between the light and the screen, the comparison light being a beam reflected from the same light on to the screen.

Owing to the fact that photographic deposit scatters light more or less, the opacities measured by this plan were slightly greater than was shown when such opacities were to be used for contact printing. The final plan adopted by Abney was to place the part of the plate carrying the deposit to be measured behind a screen

Fig. 4.

constructed as above. C D (fig. 4) is a dull black card with an aperture cut in it which may be of any desired shape. This aperture was covered with transparent paper, as was also a portion B, the same size as A, but pasted on the black card itself. Light thrown from behind A would be matched with light thrown on to B from the front when a rod in the path of this last beam was made to prevent this light falling on A. When a portion of a plate bearing a deposit was placed behind and close to A, the light thrown on B had to be diminished by the sector till the two squares appeared equally bright and the aperture of the sector was noted and compared with t at required when the deposit was removed.

With this screen accurate measures of printing densities can be made, and it can also be used in the determination of the comparative photographic brightness of the light issuing from different objects. For instance, the relative brightness of the different parts of the corona as seen in a total eclipse can be readily determined if a “time scale” of gradation is impressed on the plate on which it is taken. Both scale and streamer can then be enlarged optically and thrown on the part of the screen A. The measures of the streamer densities can then be directly compared with the densities of the scale and the relative “photographic” brightness of the different parts of the streamer be ascertained by comparison with this scale also.

The same method of measurement was adopted in ascertaining quantitatively the sensitiveness of the spectrum of ordinary plates and of plates in which dyes are present The figures on Pl. IV show reproductions of plates which were exposed to the spectrum. No. 1 is a continuous spectrum taken with the electric light; no. 7 is an impressed continuous spectrum; no. 8 shows the bright lines of metals, no. 3 the line spectrum of volatilized lithium and sodium to indicate the position of the spectrum colours. Nos. 4 and 2 are the absorption and fluorescent spectra of eosin No. 5 is the graduation scale formed by a bromogelatin “Seed” plate stained with homocol, a cyanine derivative sensitive to the red; no. 6 is a similar scale formed by an unstained plate The small numbers placed below the different bands show an empiric scale which is made to apply to each of them. The first step is to measure

Empiric Scale af the spectrum
Fig. 5.

the opacity of the gradation scale, next the opacity of the continuous spectrum at the various numbers of the empiric scale, and also the opacity of the other bands at the same scale numbers. The continuous spectrum will give the sensitiveness of the plate to the different parts of the spectrum when the measures of its different opacities are compared with those of the scale of gradation, and a curve of sensitiveness can be plotted from these com arisons. It is evident that the measures of the other two bands will give us information as to the fluorescence and the absorption of the eosin. Fig. 5 shows the curve of opacity of the image of the spectrum at its different parts, and also the curve of sensitiveness of the plate to the different parts of the spectrum. This last is derived from a comparison of the measured densities with those of the gradation scale.

Measurement of the Rapidity of a Plate.—The first attempt that was made to ascertain the rapidity of a plate was by Abney (Phil. Mag. 1874), who demonstrated that within limits the transparency of deposit varied as the logarithm of the exposure. The last formula has been accepted for general use, though it is believed that It is not absolutely correct, though very approximately true and sufficiently near to be of practical value. This belief is based on the further researches described below.^[1]

In 1888 Sir W. Abney pointed out that the speed of a plate could be determined by the formula T=E-µ(log E+C)², where T is the transparency, E is the exposure (or time of exposure Χ intensity of light acting) and C a constant. If the abscissae (exposures) are plotted as logarithms, the curve takes the same form as that of the law of error, which has a singular point, a tangent through which lies closely along the curve and cuts the axis of Y at a point which has a value of 2/√E. If the total transparency be unity, this ordinate has a value of 1⋅212, the singular point having a value of 0⋅606. The ordinate of the zero point of the curve will be where the tangent to the singular point cuts the line drawn at 1⋅212. The difference between the measurements of this zero point for two kinds of plates (i.e. C in the formula) from the points in the abscissae marking the same exposure, will give the relative sensitiveness of the two plates in terms of log x². In 1890 Hurter and Driffield (Journ. Soc. Chem. Ind. Jan. 19, 1891) worked out a less empirical formula connecting the exposure E with the density of deposit, which in an approximate shape had the form D=𝜆log(E/i), where D is the density of deposit (or log i/T),i the “inertia” of the plate, T the transparency of the deposit. In the customary way a small portion of a plate was exposed to a constant light at a fixed distance and for a fixed time, and another small portion to the same light for double the time, and so on. By measuring the densities of the various deposits and constructing a curve, a large part of which was approximately a straight line, it was found possible, by the production of the straight portion to meet the axis of Χ, to give the relative sensitiveness of different plates by the distance of the intersection from the zero point L. (See also Exposure Meters, below, under § 1, Apparatus)

Effect of Temperature on Sensitiveness.—In 1876 Abney showed that heat apparently increased, while cold diminished, the sensitiveness of a plate, but the experiments were rather of the qualitative than the quantitative order. In 1893, from fresh experiments,^[2] he found that the effect of a difference in temperature of some 40° C. invariably caused a diminution in sensitiveness of the sensitive salt at the lower temperature, a plate often requiring more than double the exposure at a temperature of about -18° C. than it did when the temperature was increased to +33° C. The general deduction from the experiments was that increase in temperature involved increase in sensitiveness so long as the constituents of the plate (gelatin, &c.) were unaltered. Sir James Dewar stated at the Royal Institution in 1896 that at a temperature of -180° C. certain sensitive films were reduced in sensitiveness to less than a quarter of that which they possess at ordinary temperatures. It appears also, from his subsequent inquiry, that when the same films were subjected to the temperature of liquid hydrogen (-252° C.) the loss in sensitiveness becomes asymptotic as the absolute zero is approached. Presumably, therefore, some degree of sensitiveness would still be preserved even at the absolute zero.

Effect of Small Intensities of Light on a Sensitive Salt.^[3]—When a plate is exposed for a certain time to a light of given intensity, it is commonly said to have received so much exposure (E). If the time be altered, and the intensity of the light also, so that the exposure (time Χ intensity) is the same, it was usually accepted that the energy expended in doing chemical work in the Him was the same. A series of experiments conducted under differing conditions has shown that such is not the case, and that the more intense the light (within certain limits) the greater is the chemical action, as shown on the development of a plate. Fig. 6 illustrates the results obtained in three cases. The exposure E is the same in all cases. The curves are so drawn that the scale of abscissae

1. Those applicable to the correction of star magnitudes as determined by photography have been verified and confirmed by Schwarzchild, Michalke and others.
2. Abney, Proc. Roy Soc. 1893.
3. Abney, Proc. Roy. Soc. 1893, and Journ. Camera Club, 1893.