OF ARTS AND SCIENCES. 279 



On standing twenty-four hours this gave hright yellow and afterwards 

 very beautiful glittering orange-red crystals. The two kinds of crystals 

 redissolved gave with mercurous nitrate a pale yellow flocky precipitate 

 which on boiling and standing became bright yellow and highly crys- 

 talline. The least soluble crystals were also redissolved separately, 

 and after twenty-four hours crystallized in beautiful prehnitic group-. 

 The orange mother liquor from these crystals gave the reaction with 

 mercurous nitrate mentioned above. It seems probable from the above 

 that 14: 6 molybdate of ammonium gives at least two distinct s;dts. 



3. With 24 : 2 sodic phospho-tungstate. A solution of this salt also 

 dissolved the Samarskite oxides very readily on boiling, giving a fine 

 orange-red solution which soon deposited an abundance of yellow 

 needles. These dissolved readily in hot water to a yellow solution 

 with an orange tint. Yellow needles quickly formed in abundance. 

 After an hour the still slightly warm mother liquor was poured off and 

 allowed to stand. Two kinds of crystals separated, — very distinct 

 rather large granular orange crystals in much the larger quantity, and 

 very small granular yellow crystals differing much from the last in 

 appearance. The two kinds of crystals were dissolved together in hot 

 water, and after a time gave flocky masses of yellow crystals. 



I did not succeed in obtaining the same results with other prepara- 

 tions of Samarskite oxides. This will not surprise those who have 

 worked with the rare earths, and who have noticed the difference in 

 the reactions which depends upon difference in the proportions of the 

 mixed oxides. It has, I believe, escaped notice that the same occurs 

 with mixtures of the different metals of the platinum group as long 

 since shown by Claus. 



The compounds with molybdic oxide and with phospho-molybdic 

 and phospho-tungstic acids may prove to be only salts of the earths, 

 and not of complex acids. They appear to deserve further attention as 

 means of separation. As the minerals belonging to the Bame group 

 with gadolinite resemble each other very closely in their physical char- 

 acters, it is possible that the yellow oxides above mentioned were not 

 prepared from Samarskite, but from some other mineral. 



Analyses 1 to 46 and from 09 to 122 inclusive were made by Mr. 

 Edward L. Smith; from 47 to 08 inclusive by Mr. Win. .!. Kar-lakc. 

 My grateful acknowledgments are due to both. 



Newport, R. I., July 30, 1893. 



