OF ARTS AND SCIENCES. 277 



true for neutral nitrates from Samarskite, and from the cerite and 

 mouazite residues already mentioned. The nitrate of roseocobalt must 

 be in excess. I have already stated that the action of the sulphates 

 of roseocobalt and luteocobalt upon the sulphates of the earths closely 

 resembles that of the alkaline sulphates. The advantage of using 

 the cobaltamines consists, in part, in the fact that the double sul- 

 phates of these and the earths are highly crystalline and exceptionally 

 well defined, and that they are in some cases at least very much less 

 soluble than the alkaline double sulphates. The chief disadvantage is 

 that the cobaltamines must be specially prepared for use, and that the 

 most valuable of them — the sulphate of luteocobalt — 'is not easy to 

 prepare in quantity and in a state of purity. Professor Morris Loeb 

 has however found that sulphate of roseocobalt may be converted into 

 sulphate of luteocobalt by heating with strong ammonia water under 

 pressure, as for instance in sealed tubes ; and as the sulphate of roseo- 

 cobalt is easily prepared, this process is perhaps the best. 



A solution of sulphate of luteocobalt gives a very insoluble yellow 

 crystalline precipitate with sulphate of thoria, Th(S04)2. It gives 

 also slightly soluble precipitates with uranic sulphate, UO2SO4, and 

 with a solution of ferric alum which has undergone dissociation by 

 solution. This last precipitate appears to have the formula 



FeoO . (SOi)., + Lco(S04)3. 



It is my hope to be able to return to the subject in greater detail. 



Relations of the Oxides to Lactic Acid. — A portion of the oxides 

 obtained from Samarskite by Dr. Shapleigh after the cerite oxides had 

 been separated by sodic sulphate was boiled with pure lactic acid, and 

 gave an amethyst-red solution. On standing, this solution gave two 

 kinds of crystals, which were very distinct and well defined. These 

 were beautiful red flat prisms, and distinct bright yellow granular crys- 

 tals. The quantity was too small to permit a more thorough exam- 

 ination, and I did not obtain the same result a second time with other 

 Samarskite oxides. In one experiment, however, the solution of the 

 oxides was deep orange, and after a time deposited crystals with a fine 

 orange color. 



The lactates of the cerite earths and of the Samarskite earths which 

 have not been treated with potassic or sodic sulphate give beautiful 

 white feathery crystals, which dissolve with difficulty in hot water. 



Relations of Merciirous Nitrate and Mercuric Oxide to Cerite Earths. 

 — A solution of mercurous nitrate gives in general no precipitate with 

 neutral nitrates of the cerite earths. In one experiment, however, in 



