542 
Proceedings of the Royal Society of Edinburgh. [Sess. 
before, the nitrate by E. Miiller * for c = *053 but only for the middle of 
the spectrum, and the sulphate by Nichols and Spencer, f E. Muller J and 
A. S. Russell and myself. § As this formed a means of testing the accuracy 
of my work, I plotted the figures for the sulphate in the following diagram. 
Nichols and Spencer give only four points, and the accuracy of their work 
is probably not intended to be great ; I have therefore not used it. M tiller’s 
figures are for c = ’ 211, and are shown as black discs. Our former results, 
which are for c = # 408, are shown as xs, and the figures recorded in this 
paper are plotted as o s. || The agreement is satisfactory. Our former 
results were, of course, made with the same instrument, but during the 
interval it has been taken to pieces once or twice. 
* “Ueber die Lichtabsorption wassriger Losungen von Kupfer und Nickelsalzen,” Ann. 
d. Phys. (4), xii. p. 767, 1903. 
t E. L. Nicbols and Mary C. Spencer, “The Iniiuence of Temperature upon the 
Transparency of Solutions,” Phys. Rev., ii. p. 344 (1895). 
f “ Untersuchungen liber die Absorption des Lichts in Losungen,” Ann. d. Phys . 
(4), xxi. p. 575, 1906. 
§ “On a Question in Absorption Spectroscopy,” Proc. Roy. Soc. Edin., xxix. p. 68 
(1908). 
|| In the former paper the values of c, for which the different values of A are determined, 
were not given. They are, for copper sulphate ’226, cobalt chloride c=’452, potassium 
dichromate *039, and uranyl nitrate *9906 (last eight points 4 this strength). 
