On Abnormal Figures in Photographic Impressions. 107 



points are of course the reciprocals of those with respect to the 

 harmonic relation of two lines, and do not need to be separately 

 stated. 



The preceding theorems are useful in (among other geome- 

 trical investigations) the porism of the in-and-circumscribed 

 polygon. 



2 Stone Buildings, 

 July 9, 1853. 



XV. On the Cause of the Occurrence of Abnormal Figures in 

 Photographic Impressions of Polarized Rings. By G. G. Stoke s, 

 M.A.j F.R.S., Lucasian Professor of Mathematics in the 

 University of Cambridge*. 



THE object of the following paper is to consider the theory 

 of some remarkable results obtained by Mr. Crookes in 

 applying photography to the study of certain phenomena of 

 polarization. An account of these results, taken from the Journal 

 of the Photographic Society, is published in the last Number of 

 the Philosophical Magazine f. 



In the ordinary applications of photography, certain objects 

 and parts of objects are to be represented which differ from one 

 another in colour, or in brightness, or in both, according to the 

 nature of the substances, and the way in which the lights and 

 shadows fall. In the photograph the objects are represented as 

 simply light or dark. Inasmuch as the photographic power, 

 in relation to a given sensitive substance, of a heterogeneous 

 pencil of rays is not proportional to its illuminating power, the 

 darkness of the objects in a negative photograph is not propor- 

 tional to, nor even always in the same order of sequence as, their 

 brightness as they appear to the eye. Still, the outlines of the 

 objects and of their parts are faithfully preserved. For although 

 it is conceivable that two adjacent parts of an object, which the 

 eye instantly distinguishes by their colour, should reflect rays of 

 almost exactly equal photographic power in relation to the par- 

 ticular sensitive substance employed, so as to be absolutely un- 

 distinguishable on the photograph, or on the other hand, that 

 two parts of an object between which the eye can see absolutely 

 no distinction should yet come out distinct on the photograph, 

 the conditions which would have to be satisfied in order that the 

 forms of a set of objects, suppose coloured patterns, or a paint- 

 ing of the rings of crystals, should be changed in this way by 

 the substitution of one set of outlines for another are so very 



* Communicated by the Author, 

 t Vol. vi. p. 73. 



