144 REPORTS ON INVESTIGATIONS AND PROJECTS. 



salts with its chloride and nitrate. The only double salt obtained 

 with the purpureo cobalt-amine was with the berzelium nitrate. The 

 crystals were of a beautiful deep violet or purplish-red color. 



The luteo cobalt gave a fine yellow microscopic crystalline pre- 

 cipitate with carolinium sulphate which was very insoluble. The 

 roseo salt was obtained in well-defined crystals of a light-reddish tint. 



The purpureo salt gave no precipitate with berzelium sulphate, as 

 noted. 



The original thorium-sulphate solution on treatment with luteo 

 cobalt-amine gave at first the yellow crystals characteristic of the 

 carolinium sulphate and then a second lot of large red crystals sim- 

 ilar in appearance to the double salt obtained with berzelium chloride. 

 While a separation could be made by taking advantage of the rate 

 of formation of these compounds, so far it has not been done with 

 perfect satisfaction. 



An atomic weight determination of the fraction thus had was made 

 and the value 216 obtained. This showed an approach to 212 of 

 berzelium. 



The type-formula of the double luteo salts was obtained by analysis 

 as follows : 



2{ 2[Co(NH 3 ) 6 Cl,] +3R(S0 4 ) 2 }■ + 3H t O 



For the roseo salt, 



2<j 2[Co(NH s ) 5 Clg] +3R(SOJ, )■ + 9H 2 



R = Original thorium =232.5 ; Carolinium = 258 ; Berzelium = 212. 



The solubility of double luteo salts in water was determined as 

 follows : 



Double salt of berzelium 1 : 2,000 parts. 



carolinium 1 : 4,500 parts. 



original thorium 1 : 5,400 parts. 



new thorium 1 : 5,400 parts. 



There was no marked difference in solubility at ioo° C. 



A NEW METHOD FOR SHOWING THE COMPLEXITY OF THORIUM. 

 Charlks Baskerviixe with E. E. Randolph. 



Although convinced from numerous chemical reactions of the 

 distinct differences of the easily volatile berzelium chloride and the 

 less volatile carolinium compound, it appeared a possibility that the 

 former of these two bodies might be a readily volatile oxychloride. 

 The oxides had been used in the preparation of the chlorides. It 

 could also be imagined that a thorium compound of a different degree 



