i8o SCIENCE PROGRESS 



than the blue and red stars just considered, will not in general 

 produce equal sensations. For ordinary emulsions, Ross finds 

 that if the photographic sensations are equal for a certain 

 exposure-time for stars of types A and K, then a tenfold increase 

 in exposure will result in a magnitude difference of 0*26, the 

 red star being estimated the fainter. 



To avoid such errors, it is desirable to use the horizontal 

 portion of the gamma wave-length curve, and this is possible 

 by matching each emulsion with a suitable filter. It is further 

 found that orthochromatic or panchromatic plates are more 

 suitable for accurate photometry than are ordinary or blue- 

 sensitive plates. 



The cause of this behaviour of the photographic plate has 

 been investigated by Ross. He finds that it is to be attributed 

 partly to a difference in the penetration of radiations of different 

 wave-lengths into the film. The penetration in the case of 

 blue light is much less than for light of longer wave-length, and, 

 other things being equal, the contrast is directly proportional 

 to the penetration. The low penetration for short wave-lengths 

 is due to absorption and scattering in the emulsion. 



The second paper of the series is entitled, ' ' Image Contraction 

 and Distortion on Photographic Plates " {ibid., 52, 98, 1920). 

 If image-diameter is plotted against the logarithm of the 

 exposure-time, the shape of the resulting curve depends upon 

 the size of the geometrical or threshold image of the object. 

 For short exposures, the diameter has its geometrical value, 

 but following this induction period there is a lag or even a drop 

 in diameter before the normal growth of the image with exposure 

 ensues. The point of maximum lag occurs with a diameter of 

 the threshold image of about 5 mm. It follows that there 

 exists a contracting influence which at first prevents the normal 

 growth of the image. 



This phenomenon, on investigation, was found to be due to 

 a contraction of the photographic image during drying. A 

 test-object, consisting of a large circular hole in a piece of 

 cardboard surrounded by a series of smaller holes, was photo- 

 graphed, and the photographs after development were measured 

 both wet and dry. A plug could be inserted in the central 

 hole, so that exposures could be made with or without the 

 central image, and the influence of the latter on the position 

 of the adjacent images determined. In this way it was found 

 that, with the large central image absent, the measures with 

 the plate wet or dry agreed, but that, with the central image 

 present, there may be a contraction of the central image during 

 drying accompanied by a drawing together of the star images 

 near its edge, the latter effect decreasing with increase in 

 distance from the edge of the central image. The magnitude 



