ASTRONOMICAL PHOTOGRAPHY 755 



the tray during development reduces the trouble to a minimuni, as in so doing a fresh 

 supply of developer is constantly brought into the region of each star image reduced. 

 The Eberhard effect when present is particularly troublesome in making "back- 

 ground" readings in the thermoelectric method of measuring a photographic plate, and 

 presents similar difficulty whenever measured diameters are involved. 



If photographic magnitudes determined from one plate are to be compared with 

 similar measurements on control plates, it is essential that both plates be developed 

 together in the same traji- and likewise receive the same treatment in fixing, washing, 

 and drying. 



There are certain characteristics of the photographic plate which the celestial 

 photographer should constantly keep in mind. Some of these greatly affect the 

 sensitivity of the emulsion. 



Experience shows that the sensitivity of photographic plates gradually deteriorates 

 with storage. After development this can be noted qualitatively by a certain degree 

 of fogginess, especially near the edges of the plate. If there is a long delay between 

 exposure and development, this change in sensitivity is well marked. Dark areas 

 tend to become darker and light areas fainter with delay between exposure and devel- 

 opment. One might say that the contrast is materiallj'- increased by delaying the 

 development. In celestial photography, however, this means that faint stars or other 

 celestial objects which might have been near the threshold value of the plate and 

 therefore just detectable upon immediate development may be lost if the time between 

 exposure and development is unduly prolonged. King states that on such plates as he 

 has tested a 15 per cent change has been noted after a month's delay. In one case, 

 where an interval of 9 months elapsed between exposure and development, a change of 

 80 per cent was noted. In terms of stellar magnitudes, differences as great as a half a 

 magnitude have been found in results, depending upon whether the magnitude scale 

 was based on the brighter or the fainter images on the plate concerned. 



Another element affecting the sensitivity of the plate is the amount of humidity 

 present during exposure. Even a small amount of moisture decreases the sensitive- 

 ness of a photographic plate seriously. When making exposures for photometric 

 purposes, if comparisons are to be made with a standard light or a comparison star, it 

 is desirable that a control exposure be made both at the beginning and at the end of 

 any series of exposures for photometric purposes. Almost as important as the degree 

 of humidity is the air tjCmperature to which the plate is exposed. With the lowering 

 of the temperature, a plate will in general show fainter stars for a given exposure time. 

 If a series of exposures is made to determine the variation of brightness of a variable 

 star and the temperature is falling, an allowance must be made for this change in 

 temperature during the series of exposures. Usually this is done by comparing the 

 images of other stars in the field whose light is regarded as constant. It is always 

 desirable that plates taken from the darkroom should be in the plateholder and exposed 

 to the temperature of the telescope several hours before the exposure is actually to be 

 naade. 



Spectrographic Measurements. — The problem of nieasuring spectrograms in 

 astronomical work is not very different from that of measuring spectrographic plates 

 from laboratory sources. The determination of line positions on a linear scale is made 

 with a suitable comparator. The reduction of scale ratings to wavelengths in the case 

 of prism spectra is usually effected by the use of the well-known Cornu-Hartmann 

 formula 



X = Xo ^ (15) 



s + a 



in which s is the measured distance from the line of reference, while X, a, and b arc 



