56 Dr. Thomas Thomson on 



a fracture like a drop of gum, which by pulverization be- 

 comes pink-red, shew that it is not madder-purple, which 

 is an orange-yellow crystalline powder, dissolving in potash 

 and forming a cherry-red solution ; therefore, the violet 

 colour described by Claubry and Persoz, indicates the in- 

 termixture of madder- red. Under the name Xanthine or 

 madder-yellow, Kuhlmann has described a constituent of 

 madder, which from its properties must be a mixture of 

 madder-orange and madder-yellow. It dissolves readily 

 in water, and forms with mordanted cotton a Pomeranian 

 yellow. This easy solubility indicates madder-yellow, and 

 the Pomeranian colour the madder-orange, and was re- 

 marked in separating the latter from madder-yellow, when 

 brought in contact with cotton impregnated with alum. 

 Madder-yellow so prepared, dyes no longer orange. The 

 pure madder-purple was described by Runge in 1823. 

 C To be continued. J 



Article VI. 



On the method of determining the proportions of Potash and 

 Soda, when the two alkalies are mixed together. By 

 Thomas Thomson, M. D., F. R. S. L. and E., &c, Regius 

 Professor of Chemistry in the University of Glasgow. 



It is no uncommon thing to meet with minerals which con- 

 tain both potash and soda as constituents. This is the case, 

 for example, with glassy felspar, couzeranite, &c. The 

 method of separating the two alkalies from each other, in 

 such cases, is that first pointed out by Dr. Wollaston. All 

 the other constituents of the mineral being separated, the 

 potash and soda are united to muriatic acid, or converted 

 into chlorides of potassium and sodium. These chlorides 

 being dissolved in water, are mixed with a solution of chlo- 

 ride of platinum. The mixture is evaporated to dryness in 

 a gentle heat, and then digested in a sufficient quantity of 

 weak alcohol. The chloride of sodium, and any excess of 

 chloride of platinum that may have been added are dis- 

 solved, while the potassium-chloride of platinum remains 

 undissolved. Separate it by the filter, wash it and dry it; the 

 potash contained in the mineral amounts to t^ths, or 0-23 



