44 
Proceedings of the Royal Society of Edinburgh. [Sess. 
which touch one another sharply on the photographic plate. The position 
of the arc was found which gives equal intensity of these spectra all over, 
a cell with the solution to be examined was placed in the path of one of the 
beams, a similar cell with water placed in the path of the other, and a 
series of exposures made for different positions of the arc. Then the 
positions of equal density on the plate were determined, and A calculated 
as in the former paper. The exposures were 30 seconds. 
Owing to the obliquity of the mirrors and consequent difference of focus 
for horizontal and vertical lines the sharp edge H of the rhomb was in 
focus 3 cm. in front of the slit. Considerable difficulty was found in 
obtaining a suitable diffuse reflector. Plaster of Paris was tried at first, 
but did not reflect the ultra-violet at all. Twelve other substances were 
tried, and an opaque fused silica finally selected. 
This method, although more convenient than that described in the 
fourth paper, did not prove as accurate owing to the iron arc not giving 
out the same quantity of light on both sides, 10 per cent, variations of the 
