38 STUDY OF ABSORPTION SPECTRA 



of the 14 solution-curves given for hydrated salts show clean-cut evi- 

 dence of a shift toward the red. This would, of course, cause the solution 

 to be less opaque as we pass into and through the centers of the bands, 

 and more opaque as we pass out of the bands and reach the points of 

 greatest transmission. It is difficult to explain why the lju solution 

 bands for magnesium nitrate and zinc nitrate (figs. 9 and 13) are more 

 opaque at the centers of the bands, especially as there is fairly good evi- 

 dence that a shift towards the red has taken place. The nitrates have a 

 widely different molecular volume from any of the other salts studied. 

 Their molecular volumes are very large. This may in some way account 

 for the peculiarities in the transmission curves of the nitrates. The 

 sulphates have small molecular volumes and in some manner produce 

 a concentration in the solvent. The transmissions at the centers of 

 the I/j. bands for solutions of zinc and magnesium sulphates are much 

 higher than for any of the other salts studied by us (figs. 5 and 7). 

 Solutions of sodium nitrate show that the nitrates probably have a 

 freer vibrating system than even the pure solvent. Solutions of this 

 salt show at the centers of both bands greater absorption, and at the 

 tops of the bands higher transmission than the pure solvent. More 

 attention should be given, in work of this character, to the volumes 

 and masses of the vibrating systems, which undoubtedly vary widely 

 with the addition of different salts. 



It has been seen that in only 4 out of 14 curves for the hydrated 

 salts is there any evidence of a shift of the solution curve towards the 

 red. In our opinion the chief factor which makes the aqueous solution 

 more transparent than a depth of water equal to the water in the solu- 

 tion is that hydration exists in the solution. It is quite certain now 

 that in the solutions which we studied, much more than half of the 

 water was combined with the dissolved substance. It seems almost a 

 necessity that this would alter the vibrational frequency of the absorb- 

 ing electrons or systems. The character of the transmission curves 

 seems to justify the conclusion that water of hydration has less power 

 to absorb light than pure uncombined water. We can conceive of no 

 other rational explanation which will interpret satisfactorily our results 

 for hydrated salts. That water of hydration is less opaque to light 

 than free water seems, from our work, fairly certain. In this way alone 

 is it possible for us to explain satisfactorily the transmission curves 

 for hydrated salts. 



A question of importance is why, for non-hydrated salts, is the 

 transmission curve for the solution always below the transmission 

 curve for the solvent, not only at the centers of the bands, but for 

 most of the regions of the spectra which we studied? This is exactly 

 the reverse of the relations already pointed out for the hydrated salts. 

 Just before the centers of the bands are reached, passing in the direc- 

 tions of the longer wave-lengths, the solution of the non-hydrated salt 



