28 ABSORPTION SPECTRA OF SOLUTIONS. 



absorbs all light of shorter wave-length than X 3000. The absorption 

 recedes gradually with addition of water, and a little more rapidly at first, 

 as is apparent from the curvature of the edge of the band nearest the scale. 

 The solution containing 11 per cent of water transmits light of a wave- 

 length as short as X 2560. The green band has very nearly the same inten- 

 sity as it had in the methyl alcohol solutions just described, and narrows 

 with addition of water to about the same extent. Measurements give its 

 center for the solution containing no water at X 5220, and for the solution 

 containing 11 per cent of water at X 5200. 



The absorption in the red is of a general character, showing only faint 

 indications of bands. The whole spectrum of the solution containing no 

 water is weak throughout the entire red region, and shows weak bands 

 superposed upon the general absorption, having their centers at about 

 X 6150 and X 6730. These regions correspond roughly to the minima of 

 sensibility of the panchromatic plates ; therefore, due allowance must 

 be made for this in studying the spectrogram. The absorption decreases 

 rapidly with increase in the amount of water, being practically absent in 

 the solution containing 2 per cent. All solutions containing a greater 

 percentage of water are completely transparent to the red as far as the 

 limit of sensibility of the plates used. 



COBALT BROMIDE IN ACETONE WITH WATER. (See Plate 18.) 



The concentration of the colored salt throughout was constant and 

 equal to 0.007 normal. The successive percentages of water, beginning 

 with the solution whose spectrum is nearest the numbered scale, were 

 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, and 15. 



The first seven solutions were of various shades of blue, from dark to 

 light-blue. The eighth, ninth, and tenth were light-bluish to bluish-green, 

 after which the solutions were almost colorless, a very faint pinkish tinge 

 only being noticed. 



The depth of absorbing layer used was 2.0 cm., and the exposures to 

 the Nernst lamp and spark lasted for 1 and 3 minutes, respectively, the 

 slit being, as usual, 0.01 cm. wide. 



As has been repeatedly observed, acetone absorbs all wave-lengths 

 shorter than about X 3300 for the thickness of layer here used. The edge 

 of the band is unusually sharp, which is, however, not the case for the 

 first eight strips of Plate 18. In fact the strip nearest the scale just barely 

 records the line at X 3320, the shading from this point towards X 3SOO being 

 considerable. This shading is undoubtedly due to the effect of the colored 

 salt. It decreases with addition of water, but is still easily noticeable in 

 the solution containing 15 per cent of water. 



Measurements on the edges of the band in the red settings being made 

 on the limits of transmission, give the following: 



Solution which contains no water, ?- 5600 to beyond / 7400 

 Solution containing 1 p. ct. of water, A 5680 to beyond / 7400 

 Solution containing 2 p. ct. of water, ? 5750 to beyond / 7400 

 Solution containing 3 p. ct. of water, ? 5850 to ? 7400 

 Solution containing 4 p. ct. of water, ? 5970 to / 7350 

 Solution containing 5 p. ct. of water, ? 6070 to ?. 7250 

 Solution containing 6 p. ct. of water, absorption band broken up. 



