218 Mr. A. W. Flux on the 



and perpendicular to the plane of incidence, in two straight 

 lines, which cross each other without intersecting. 



We shall refer to these lines as the " Principal" line and 

 the " Transverse " line. 



(2) The rings are the curves in which this ruled surface is 

 intersected by a set of coaxial cylinders, which are oblique 

 cylinders on circular bases. 



In the case of the transmitted system the rings are not 

 rigidly coaxial, though for ordinary angles of incidence the 

 defect could only be detected by means of very refined in- 

 struments. 



(3) The rings are symmetrical with regard to the first of 

 the two planes mentioned in (1), both in shape and distinctness. 



(4) The principal line being inclined to the surfaces of the 

 plate of glass used in producing the rings, after the ring of a 

 determinate order, every ring will lie partially above the upper 

 surface of the plate. Similarly every ring beyond another 

 determinate limit will lie partly below the lower surface. 



(5) The points of the principal line are always clearly 

 defined. In all other directions the clearness varies with the 

 incidence, decreasing as the angle of incidence increases, in . 

 general. 



Every ring whose order does not exceed a certain limit, de- 

 terminate for each angle of incidence, will have, however, a 

 pair of points quite clearly defined. These points lie sym- 

 metrically on the side of the rings furthest from the incident 

 light. Further, the rings are always more clearly defined on 

 this side than on the other. 



(6) Any point in a certain line intersecting the principal 

 and transverse lines, and lying between these lines, may be 

 taken as the central spot, which is therefore, to that extent, 

 indeterminate. 



The analytical methods of this paper are, for the most part, 

 derived from an article in Band xii. of Wiedemann's Annalen 

 der Physik unci Chemie, by Herr Wangerin. Experimental 

 evidence supporting the theory may be found in the first 

 article in that volume. 



The results attained by Wangerin are in all respects iden- 

 tical with those of this paper. I have endeavoured to profit 

 by some changes in Herr Wangerin's method, developed in 

 later articles, and to modify these and combine them so as to 

 derive a method which seemed to me to be satisfactory. 



The results for transmitted light have been given by Herr 

 Gumlich in Band xxvi. of Wiedemann's Annalen. 



