4 PROCEEDINGS OP THE AMERICAN ACADEMY 



The formula, Coo(NH3)jQ(N02)2Br^, requires cobalt 16-86, bromine 

 45-71. I did not succeed in finding the sulphate of the octamin series 

 among the products in the single experiment which I made with the 

 above-mentioned mixture. If present at all, its relative quantity must 

 have been small. Tlie formation of sulphate of xanthocobalt is easily 

 explained by the equation : — 



2CoSO,4-10NH.,4-2KNO2 + OH.,4-O = 

 Co,(NH3),„(NO,),(SO,), + 20KH; 



but it is not easy to see why the presence of amraonic sulphate should 

 determine the jDroduction of sulphate of xanthocobalt in place of the 

 sulphate of the octamin series. 



With these preliminaries I pass to the description of the salts of the 

 new octamin series. These salts as a class greatly resemble those of 

 xanthocobalt, but are rather more stable. They have a fine sherry- 

 wine color, are usually comparatively insoluble in cold, and are dis- 

 solved with difficulty even by boiling water. The solutions when 

 neutral are decomposed by boiling, ammonia being evolved and a black 

 powder precipitated. The addition of a small quantity of acetic acid 

 serves to prevent the decomposition in hot solutions. Mineral acids, 

 even in small quantity, usually produce more or less decomposition on 

 heating. The salts crj^stallize with remarkable facility, resembling in 

 this respect the salts of luteocobalt, which are, however, much more 

 soluble. As the octamin salts are easily prej^ared in quantity, they 

 may hereafter be found to possess some value as means of investigation 

 from their remarkable power of crystallization. 



Sulphate. — Of all the salts of this series the sulphate is that which 

 is most easily prepared in quantity and free from other products. The 

 general method of preparation has been already pointed out. The 

 mixture of cobaltic hydi-ate and crude sulphate is to be boiled with very 

 dilute sulphuric acid, filtered, and the solution allowed to stand for a 

 few hours, when the sulphate sei3arates almost completely, in conse- 

 quence of its insolubility in cold water. The mother liquor contains a 

 Lirge quantity of cobaltic sulphate, and traces of the new salt together 

 with potassic and ammonic sulphates. A second crystallization gives a 

 perfectly pure salt. In large crystals the salt has a dark wine-red 

 color, like the salts of xanthocobalt. It usually separates from hot 

 concentrated solutions in small, very brilliant yellow scales, which, under 

 the microscope, appear to belong to the quadratic system. The sul- 

 phate is remarkable for its insolubility. Cold water dissolves a very 

 small quantity, the solution taking a golden-yellow color. Even in 



