66 Scientific Proceedings, Royal Dublin Society. 



flask connected with a reflux condenser. Some pieces of linen, wool, 

 silk, and cotton were washed with soap and water, and placed in running 

 water over night, when small dry portions were added to the solution and 

 boiled for two hours. On removal it was found that neither the wool, silk, nor 

 cotton was changed in colour ; but the linen when laid on white paper was 

 faintly blue. This test with silk- and wool-fibres serves to distinguish the 

 colour from that of a solution of aniline violet. The colour reactions with 

 acids and alkalis are also difl'erent. 



Some 5"0 c.c. of an alcoholic solution containing O'OOo grm. of solid were 

 placed in an air-tight glass weighing-bottle. This was placed in a window 

 for twenty-four days, during which time it was not exposed to more than 

 twelve hours' direct sunlight, and at the end of the period the solution was 

 almost colourless ; only a faint trace of red remained. 



The Absorption S^Jedrum of the Colouring-Matter. 



A portion of the first alcoholic solution was evaporated partly on a water- 

 bath and partly in vacuo, and repeatedly weighed till constant, the weight 

 of the colouring-matter being 0'022 grm. A preliminary examination was 

 made with a miscrospectroscope by Zeiss, conveniently fitted with a scale of 

 wave-lengths. 



A glass tube, 27 m.m. long by 10 mm. internal diameter, was cemented 

 vertically to a thin glass microscope-slide, and in this the solution was placed. 

 Thicknesses of solution greater than 10 mm. showed transmission of the blue, 

 indigo, and a portion of the violet rays, or from a little beyond the solar line 

 JPto a point half way between G and H. There was total absorption of the 

 red rays from A to beyond B ; but a narrow band of bright red light with its 

 centre about C was transmitted. Small thicknesses showed transmission of 

 the bright red rays from X 670 to A 640, and absorption from about X 640 to 

 X 490, the rays beyond F being transmitted. 



A more precise examination of the same solution was made with one 

 of Hilger's fixed deviation wave-length spectroscopes, illumination being 

 by sunlight directed on to the slit by a heliostat. 



No rays were transmitted through a thickness of 25 mm. ; through 

 smaller thicknesses, the measurements resembled those obtained with the 

 Zeiss instrument and, stated generally, the rays less refrangible than 

 X 6700 were absorbed ; the bright red rays from about X 6670 to about 

 X 6240 were fully transmitted. 



The rays from X 6100 to near F (X 4900) were absorbed ; but a feeble 

 transmission near the green 6' 6' group to beyond F was seen, with a complete 

 transmission from X 4900 to X 4000. The variations in the sunlight on the 



