AESUiiPTIuN SPECTKA OF SOLUTIONS. 39 



were adjusted so as to bj veiy nearly inversely as the concentrations. The thickness of 

 the weak solution was, however, a little less than exact adjustment required. With exact 

 adjustment therefore there would have been a still greater relative motion of the regions 

 of absorption. He states also that two other series of observations with solutions having 

 smaller differences of concentration showed a motion of the absorption region in the same 

 sense, but he does not say what the concentrations were in these cases. Knoblauch 

 ascribes this observed effect of change of concentration to a supposed decomposition of 

 the salt by the water. Possibly it may be so. But it is worth noting that the concentra- 

 tion of his strong solution was very considerable, and that, as shown above, the effect 

 which he observed is exactly the effect which is to be expected in the case of strong solu- 

 tions ; for we saw that in such cases increase of concentration will produce an opposite 

 effect to that of elevation of temperatiire. 



Knoblairch examined two solutions of picric acid, one containing 9"3 and the other 

 0'00141 grm. per cu. dm., the thicknesses being inversely as the concentrations ; and he 

 found that when allowance was made in the case of the weaker solution for the selective 

 absorption of the water, the two spectra were undistinguishable. Melde found that the 

 absorption region of the spectrum of a solution of this acid (a one-sided spectrum with 

 absorption at the violet end) was extended towards the red by elevation of temperature. 

 According to the above hypotheses, therefore, increase of concentration may be expected 

 to produce the same effect in solutions of moderate strength. But it will be noted that 

 Knoblauch's solutions were both comparatively Aveak. His result, therefore, is in agree- 

 ment with the conclusions reached above. 



Of the potassium salt of eosin, the so-called "soluble eosiu," Knoblauch examined 

 two pairs of solutions, containing respectively 6"5 and 0'65, and 026 and 0000015*7 grms. 

 per cu. dm. In the case of the stronger solutions it is not clear that their thicknesses 

 were exactly adjusted. In the case of the weaker they were. In the former he ob.served 

 one of the absorption bands of the stronger solution to be nearer the red than in the 

 weaker solution. In the latter he says the absorption spectra were the same, without 

 giving the Avave lengths of the boundaries of the regions of complete and of incomplete 

 absorption. Though he says the absorption spectra in the case of the weak solutions 

 were the same, he says also that they were so nearly identical with the spectra of solu- 

 tions of nearly corresponding concentrations of the sodium salt that the diagram given 

 for the latter may be taken also as representing the former ; and according to that diagram, 

 while the A'iolet-ward boundary of the absorption band is the same for solutions contain- 

 ing 1 and 000013T grm. per cu. dm., the red-ward boundary of the stronger solution is 

 nearer the red than that of the weaker, the red-ward boundary of the region of complete 

 absorption being, however, the same in both. On the whole, therefore, if increase of con- 

 centration has any effect other than change of intensity on the absorption of this salt, it 

 would seem to be a slight extension of absorption towards the red. 



With regard to the effect of elevation of temperature on the absorption of solutions of 

 this salt, G-. and H. Kriiss' give several series of observations. They found, by spectro- 

 scopic observations, in the case of a solution whose strength they do not specify, that on 

 changing the temperature from 20' to 40', 60 and 80' C-. the position of the point of maxi- 



1 ' Kolorimetrie,' pp. 275 and 278. 



