clear glass, which upon cooling, and also when the blast is directed on it for a long time, becomes turbid; a larger addition of soda converts it into a difficultly fusible enamel, which remains on the charcoal while the excess of soda is absorbed.

The Yttro-cerite from Broddbo decrepitates but slightly when heated in a flask. It is infusible on charcoal, but on heating it becomes first milk-white, then brick-red, but, however, always unequally colored. It does not fuse with gypsum.

It behaves to borax and microcosmic salt in a nearly similar manner to oxide of cerium, as it contains a large quantity of fluoride of cerium mixed with it.

It is acted upon by soda, tumefies a little, but is not dissolved. The greater part of the soda is absorbed by the charcoal, leaving a grey scoria behind.

If it be required to separate the yttria from Yttro-cerite, it must be done in the moist way, according to the following method; the finely powdered mineral is to be dissolved in hydrochloric acid, heated to ebullition, the solution evaporated nearly to dryness, diluted with water, and filtered in case any undissolved particles are observed. From this diluted solution, alumina, yttria, and protoxide of cerium may be precipitated by an excess of ammonia; the whole should then be warmed, and the precipitate separated by filtration. Lime is precipitated from the filtrate by oxalate of ammonia. The precipitate obtained by ammonia should be edulcorated with hot water, until the water which percolates through the filter is no longer rendered turbid by oxalate of ammonia, and digested at a moderate heat while still moist with caustic potassa, in order to separate the alumina from the yttria and protoxide of cerium. When the digestion is complete, the alkaline solution is to be diluted with water, and filtered. The alumina may be separated from the filtered solution slightly heated, by the necessary quantity of chloride of ammonium, or by ammonia from the solution, rendered acid by hydrochloric acid. In both cases, the precipitate after filtration must be well edulcorated, and then tested with a cobalt solution. The undissolved yttria and protoxide of cerium is to be well washed with hot water, dissolved in hydrochloric acid, and the

solution diluted with water. A crust of crystallized sulphate of potassa must then be placed on the surface of the liquid, and the whole allowed to stand about twenty-four hours. During this time the solution will become saturated with sulphate of potassa, with which the yttria and protoxide of cerium will combine, forming double salts; of which the one formed with yttria is soluble in a saturated solution of sulphate of potassa, while that formed with protoxide of cerium, on the other hand, is insoluble, and falls to the bottom as a white amorphous powder. After the lapse of the above mentioned time, the remaining crust is removed, the precipitate separated by filtration, and edulcorated with a saturated solution of sulphate of potassa. After the edulcoration, it is to be dissolved in boiling water, the protoxide of cerium precipitated from the still warm solution by potassa, filtered, the precipitate edulcorated, and ignited in the platinum spoon. During the ignition it is converted into sesquioxide, and becomes of a cinnamon brown color. The yttria which remains in solution is also to be thrown down from the warm solution by potassa, filtered, and ignited in the platinum spoon. It must possess a white color after the ignition; if it becomes brown, it contains sesquioxide of iron, which can be separated after the earth has been dissolved in hydrochloric acid, by the method already given for phosphate of yttria. A separate examination is required for the detection of hydrofluoric acid, which will be subsequently given.

c. Polymignite, the name of which is derived from #ovs much, and μyvow, I mix,-according to BERZELIUS, behaves in the following manner.

Heated in a bulb tube, it yields no water; it is infusible on charcoal, and does not change its color or brilliancy. Borax readily dissolves it to an iron colored glass, which, by a larger addition, acquires the property of becoming streaked with turbid lines, by which it generally appears of a brownish-yellow color. From a larger addition, it becomes turbid per se, on cooling; with tin the bead becomes of a red color, passing into yellow.

It is difficultly dissolved by microcosmic salt, the glass assuming a reddish color in the oxidating flame, which is not altered

on the addition of tin. It is decomposed by soda, becoming greyish-red, but does not fuse; it gives the same reaction as manganese on platinum foil.

In order to separate the yttria, the moist way must be had recourse to, as in the case of the first two minerals. The method of procedure is as follows:-The finely powdered mineral is to be fused with from six to eight times its weight of bisulphate of potassa, in a platinum spoon, and the fused mass digested with a sufficient quantity of water, until the whole is dissolved out of the spoon; the sulphates of the protoxides of iron, cerium, and manganese, of yttria, lime, and magnesia, are dissolved, while the titanic acid, in combination with zirconia, and a part of the protoxide of iron, remains behind; the whole must then be kept for some time at a moderate ebullition, and filtered. The protoxide of cerium may be separated from the solution by a piece of sulphate of potassa, in the manner already described under Yttrocerite. To the solution, from which the double salt has been separated by filtration, a few drops of nitric acid are to be added, in order to convert the protoxide of iron into the sesquioxide; a little chloride of ammonium should also be added, to prevent the precipitation of protoxide of manganese and magnesia, if it should contain them, and the yttria and sesquioxide of iron thrown down with ammonia. Lime is separated from the filtered solution, by oxalate of ammonia; and magnesia and protoxide of manganese, by microcosmic salt. The precipitate obtained by ammonia must be dissolved in hydrochloric acid, and the sesquioxide of iron separated from the yttria, by the method already given for the decomposition of phosphate of yttria.

d. The following has been found to be the behavior of Yttrotantalite before the Blowpipe :-Heated per se, in a bulb tube, it gives off water, and, if previously black, becomes yellow. Some specimens become mottled with dark spots, which remain unchanged in the heat; they become white on ignition, while the upper part of the glass is attacked. The water eliminated first colors brazil-wood paper yellow, and then bleaches it.

Borax dissolves it to an almost colorless glass, which, by a larger addition, becomes turbid on cooling.

L

It is at first decomposed by microcosmic salt, the tantalic acid remaining undissolved as a white skeleton, which fuses, however, in a strong blast. The black variety of Yttro-tantalite, from Ytterby, produces a glass which becomes of a light rose color by a strong reducing flame, on cooling, derived from the presence of tungsten. The dark and yellow varieties from Ytterby, become, on cooling, of a slight but beautiful green color, from the presence of uranium. The Yttro-tantalite, from Finbo and Kararfvet, gives a strong ferruginous tint, which becomes darker when the test for oxide of uranium is employed.

It is decomposed without solution by soda. It exhibits on platinum foil the presence of manganese. Traces of tin are obtained by reducing with soda and borax; the Finbo variety, however, contains so much iron that the tin cannot be recognized.

The preparation of yttria from Yttro-tantalite is attended with much difficulty. The separation can only be performed in the moist way, and safest when more than 100 milligrammes of the mineral are employed, for which larger vessels are required than the Blowpipe operator can conveniently carry about with him. e. Silicious combinations :—

1 Gadolinite, from Ytterby, Finbo, and Broddbo ;—

BERZELIUS gives the following account of their behavior before the Blowpipe. These Gadolinites are of two varieties, one of which, a, is of such a vitreous nature that it appears like black glass; the other, ẞ, on the contrary, has a splintery fracture, and not so conchoidal; the former contains no glucina, but it is very probable that the latter has some per cents.

Variety a, per se, in a bulb tube, undergoes no change, and gives off no water; if it be heated nearly to the point of fusion, it appears to glow for an instant, as if it had caught fire; it tumefies a little, and if the fragment be large, fissures appear here and there in it, and the color assumes a light greyish-green; no volatile matter is, however, given off. The same phenomena take place on charcoal; it does not fuse, but in a strong heat it becomes black on the edges.

Variety B, per se, swells out into dendritic ramifications, and

becomes white, at the same time giving off moisture. The above mentioned luminous appearance is very rarely observed. In all other points both behave alike to fluxes.

It is readily dissolved by borax, forming an intense ferruginous colored glass, which becomes of a deep bottle-green in the reducing flame.

It is difficultly dissolved by microcosmic salt. The glass assumes a ferruginous color, and the fragment employed becomes rounded on the edges, but it remains white and opaque, so that the silicic acid cannot be separated from the phosphoric acid; by this it is principally distinguished from the Kararfvet variety. Soda dissolves it to a reddish-brown semifused scoria. The variety ẞ melts into a bead with soda, when too large a quantity of the flux is not employed. None of them exhibit the slightest trace of manganese on platinum foil.

The moist way must be had recourse to, in order to separate yttria from Gadolinite. The method of procedure is as follows:

The finely powdered mineral is to be fused to a bead, with an equal bulk of carbonate of soda and borax, on charcoal, in the oxidating flame; the bead pulverized, the powder dissolved in hydrochloric acid, the solution evaporated to dryness, the dry mass dissolved and filtered from the insoluble silicic acid. A few drops of nitric acid are then to be added to the solution, and the whole heated, in order to convert the protoxide of iron into the sesquioxide, and the yttria, protoxide of cerium, sesquioxide of iron, and glucina, when present, thrown down by ammonia. Lime may be precipitated from the supernatant liquid, when present, by oxalate of ammonia. The precipitate by ammonia, after edulcoration with hot water, must be treated in the heat while still moist, with a solution of caustic potassa, in order to separate any glucina present; the alkaline solution should then be diluted with water, and the glucina separated from it by chloride of ammonium, in the way already given under the head Glucina. The residue, after edulcoration, is then to be dissolved while still moist, in a small quantity of hydrochloric acid, and the protoxide of cerium separated by sulphate of potassa, by the method given