100 Scientific Proceedings, Royal Dublin Society. 



which have the appearance of a line and a shadow beside ifc. A 

 train of four or six water prisms, no doubt would remedy this 

 by giving increased dispersion to the rays ; but a longer exposure 

 of the photographic plate would be necessary, on account of the 

 strength of the rays being diminished by reflection from the many 

 surfaces t)f quartz of which the sides of the prisms arc made, and 

 the apparatus would be complicated by the multiplicity of 

 prisms. 



For several reasons I prefer to employ quartz prisms. In order 

 to ascertain the position of the plate with regard to the ray of 

 mean refrangibility, a calculation based on the refractive power 

 of the prisnij for one of the ultra-violet lines of cadmium, as for 

 instance line 17, as well as the focal length of the lens for this 

 ray as compared with those constants for the line D, gives us 

 roughly what is afterwards determined more accurately by 

 experiment. 



With a quartz prism of 60°, the angle which that half of the 

 plate on which lies the most refrangible part of the spectrum 

 must make with the direction of the mean ray, varies between 

 19° and 21°, with lenses of 15 and 86 inches focus. The camera 

 back is therefore constructed so as ordinarily to take up this 

 position, but it is made to swing on a vertical pivot exactly 

 situated at its centre. The centre of this pivot is in the same 

 vertical plane as the centre and front surface of the focussing 

 screen, which again, as is usual, corresponds with that of the 

 photographic plate. 



It follows from this that if any particular ray situated at the cen- 

 tre of the plate be accurately focussed no alteration in the inclina- 

 tion of the plate will displace this line. The mean ray of cadmium is 

 the line 17, and the first thing to be done in adjusting the camera 

 is to place the prism at the angle of minimum deviation for this 

 ray, then the ray is brought into the centre of the focussing 

 screen and carefully focussed by a backward or forward motion 

 of the camera body communicated by the focussing screw. After 

 this the " side-swing " enables the extreme of both ends of the 

 spectrum to be brought into focus. Nothing more then remains 

 to be done but to clamp the camera-body in position. 



Details of construction of the camera for lenses of long focus. 

 — In Plates 12 to 15 the elevation and plan of this instru- 



