THEORIES OF DYEING 327 



compounds in the ordinary sense, or solid solutions or " adsorp- 

 tion compounds." The point was investigated by agitating 

 metallic oxides or hydroxides with solutions of mordant dyestuffs, 

 both in the cold and at the boiling temperature of the solvent, 

 and noting the effect of varying the concentrations of the dye- 

 stuff solutions when the amount of metallic oxide was kept 

 constant. The amount of dyestuff left in solution was determined 

 colorimetrically after removal of all suspended matter by filtra- 

 tion and curves were then constructed showing the concentrations 

 of the dyestuffs in the lakes as ordinates and concentration of 

 the dyestuff in the final solution as abscissae. In the case of the 

 reaction of Fe 2 3 on a solution of alizarin in 1 per cent. NaOH 

 solution in the cold the results obtained showed that a chemical 

 compound of the formula Fe 2 3 , 3C 14 H 8 O l is formed. On the 

 other hand the action of boiling aqueous solutions of alizarin red 

 on chromium hydroxide led to the production of adsorption com- 

 pounds, since in this case the results agreed fairly well with the 

 formula 



C n oxide 

 C dye-bath = ' n ng eq t0 3 and k to a ' 1 ' 



Further experiments, in which the action of ammoniacal and 

 alcoholic solutions of alizarin and aqueous solution of alizarin- 

 blue on Fe 2 3 and of alcoholic solutions of gallein on A1 2 3 were 

 investigated, did not lead to definite results. The curves obtained 

 end in an almost horizontal portion, but the distance of this 

 from the abscissa does not agree with any simple stoichiometric 

 formula. Moreover, by comparing these results with those 

 obtained with Fe 2 3 and alcoholic solutions of alizarin yellow, 

 cloth, red, etc., the amounts of dyestuff taken up by the same 

 weight of Fe 2 3 were not found to stand in any stoichiometric 

 relation with one another. It was also found that in all the 

 cases investigated the exact physical condition of the ferric 

 hydroxide used greatly influenced the results. 



The work of Biltz seemed to show, in so far as pure experi- 

 mentation alone could do so, that the process of dyeing really 

 was a phase in the coagulation of colloids and belonged to that 

 class of reaction known generally as " adsorption phenomena." 



The matter was taken up in 1907 by Freundlicli, who attacked 

 the problem from the standpoint of Willard Gibbs's theory of 

 surface concentration. 



