Hartley — On a Violet Colouring-Matter^ ^c. 67 



few occasions when it was available made it difficult to measure accurately ; 

 and therefore it was decided to rely on the photographs of the absorption- 

 spectra, which were taken in the following manner : — 



A weighed quantity of the dried pigment, 0-022 grm., was dissolved 

 in 5 c.c. of " absolute " alcohol and successively diluted to 10, 15, 20, 40, 80, 

 and 160 c.c. (see Curve). (The solid was not entirely dissolved by 5 c.c. of 

 cold alcohol, but was completely dissolved by 10 c.c.) 



The spectrum of the alcohol was photographed through the same thickness 

 (5 mm.). It transmitted all the rays to X 2495 strongly, and feebly to X 2196. 

 The source of light was the continuous rays and metallic lines emitted by a 

 condensed spark passing between electrodes, one of which was composed of an 

 alloy of cadmium, 15 per cent, tin, 85 per cent.; the other cadmium, 15 per 

 cent, lead, 85 per cent. 



The cell was of glass with quartz ends, giving a layer of liquid 5 mm. 

 thick. The instrument was a quartz spectrograph, photographing lines, 

 from X 7500 to X 2145 in focus on cue plate, and extending to a length 

 of more than 100 mm. 



The plates were Wratten and Wainwright's panchromatic spectrum 

 plates. 



The rays from the spark were condensed by a large quartz lens on the 

 cell, wliich was placed so that a sharp image of the spark was focussed on 

 the slit of the instrument, the exposure for each photograph being one 

 minute. 



[Diagram and Tables. 



