Miscellanies. 207 



when I became acquainted, through the medium of the scientific journals, 

 with Mosander's discovery of didymium.* It has been my good for- 

 tune to discover, in a concentrated solution of valerianic acid, the means 

 of separating the oxide of cerium in a state of purity. Valerianic acid 

 possesses a remarkable and unexpected affinity for the oxide of ceri- 

 um, inasmuch as it precipitates it abundantly from a concentrated 

 solution in nitric acid, containing at the same time oxide of didymium. 

 The yellowish-white precipitate consists merely of the valerianate of 

 the oxide of cerium, and it is only requisite to wash it well and to sub- 

 mit it to a strong heat without exposure to the air, to obtain the oxide of 

 cerium in a state of purity. This oxide is of a very pale yellow color, 

 similar to that oxide as prepared by Mosander, who however confesses 

 that he has not hitherto succeeded in discovering a method of sepa- 

 rating from each other completely the oxides of cerium, lanthanum and 

 didymium. 



The oxide of didymium remains in solution in the acid fluid from which 

 the oxide of cerium has been precipitated. A portion of the latter ox- 

 ide remains, however, associated with that of didymium ; for the vale- 

 rianates of both these metals are to a certain extent soluble in water, 

 and even more so in acidulated fluids, the oxide of didymium especially, 

 which is far more easily soluble in weak acids than that of cerium. By 

 means of valerianic acid, however, the oxide of didymium may be ob- 

 tained in a state of purity, though not so readily as the oxide of cerium. 



I have only to remark, in conclusion, that to obtain the valerianate of 

 the oxide of cerium in a state of purity from the nitric solution of the 

 mixed oxides of cerium and of didymium, this salt must be precipitated 

 by an aqueous concentrated solution of valerianic acid ; — with a soluble 

 valerianate the didymium would be thrown down at the same time, inas- 

 much as the latter, when in the form of a valerianate, is but sparingly 

 soluble in neutral solutions. It is therefore upon the ready solubility of 

 the valerianate of the oxide of didymium in acidulous solutions, and upon 

 the inferior solubility of the analogous salt of the oxide of cerium, that 

 the simple preparation of the oxide of cerium in a state of purity is based. 



3. Sillimanite and Monazite. — These two minerals which are associa- 

 ted in the localities at Norwich and Chester (Saybrook) in Connecticut, 

 have recently been found together in a quarry in Yorktown, Westches- 

 ter County, N. Y., by Mr. I. Mekul. They are there associated with mag- 

 netic iron and quartz, and the crystals of Sillimanite often penetrate this 

 mineral for several inches. Mr. M. remarks, in a letter to me dated 

 Nov. 21, that they are frequently six or more inches in length, much 



* American Journal of Science, Vol. xliii, p. 404. 



