Page:Proceedings of the Royal Society of London Vol 1.djvu/356

Rh position of every point might be determined with sufﬁcient exactness for every practical purpose.

The time required for such an examination is estimated to be about ninety-eight hours, and the labour, no doubt, is very considerable; but when the errors thus ascertained have been duly noted in a table, Mr. Lax considers the utmost paius that can be bestowed upon any instrument to he amply compensated by the conﬁdence given to every subsequent observation by means of it.

It is also proposed occasionally to obviate the effects of unequal expansion in any particular observation, by comparing the are by which any angle has been measured with several succeeding equal arcs, until the multiple exceeds the whole circumference, and thereby includes the opposite errors, which arise from this cause in different parts of a circle, and correct each other.

The author having received specimens of the Swedish mineral tan- talite, containing the metal called Tantalium, by Mr. Ekeberg, was desirous of ascertaining whether that metal might not be the same as colnmbium, which had been discovered a short time before by Mr. Hatchett; and for that purpose he procured some oxide of co- lumbium from Mr. Hatchett, and also a fragment of the mineral in the British Museum, originally analysed by Mr. Hatchett.

He describes the external resemblance to be such, that one might be taken for the other; but observes, that the columbite is rather more brittle than tantalite.

By analysis, also, he ﬁnds them to consist of the same three in- gredients; namely, a white oxide, iron, and manganese.

To separate these substances, the mineral is powdered and fused with carbonate of potash and a small proportion of horax. The iron and manganese may then be dissolved, along with the salts employed, by muriatic acid, and the oxide of columbium or tantalium remains as a white powder for further trial of its properties.

Five grains of columbite being thus treated, left four grains of white oxide; and the solution yielded three fourths of a grain of iron, and one fourth of a grain of manganese.

Five grains of tantalite, by the same treatment, left four grains and a quarter of oxide, half a grain of iron, and two tenths of a grain of manganese.

The white oxides obtained from each of these minerals appear to the author to have precisely the same properties.

They are each soluble by means of about eight parts of potash.

They are both very imperfectly soluble by means of soda.

They are both insoluble in nitric, muriatic, succinic, and acetic acids.

They are both Very sparingly soluble in strong sulphuric acid