Page:Proceedings of the Royal Society of London Vol 1.djvu/357

Rh while boiling; but they are nevertheless both perfectly soluble in oxalic acid, in tartaric acid, or in citric acid.

They are both precipitated of an orange colour by infusion of galls, but are not precipitated by that re-agent if a considerable excess either of alkali or acid prevail in the solution.

As a further agreement in their properties, it is added, that neither of them is precipitated by prussiate of potash or by hydrosulphuret of potash.

From these experiments, although a great difference which sub- sists between the speciﬁc gravities of the two minerals cannot be very satisfactorily explained, the author is satisﬁed that the Ameri- can and Swedish minerals, in fact, contain the same metal.

The instnunent here described by the author is designed to obviate the inconvenience which has been found in attempting to measure any small crystals by the instruments hitherto used for that purpose.

When a surface is so small as one ﬁftieth of an inch in breadth, it becomes extremely difﬁcult to apply the short radius of a goniometer to it with correctness. But since a surface of that magnitude may reﬂect a very brilliant light, the reﬂected ray may be employed as radius, and may at pleasure be taken of such a length that the angles of small crystals can be known with as much precision as those of the largest surfaces.

The crystal being attached to a horizontal axis, with its edge in the line of the axis, one of the surfaces is made to reﬂect some bright light to the eye; and, while the eye is retained steadily in the same place, the axis is turned till the second surface reﬂects the same light, and is consequently in the same position. The number of degrees through which the axis has turned being the supplement to the re- quired angle, the angle itself is indicated by the graduations of a circle which moves with the axis ; but the complete construction of the instrument cannot be distinctly understood without reference to a ﬁgure that secompanies the paper.

Since any inaccuracy in placing the crystal would occasion some error by parallax in this method of using the instrument, the author describes a second method, by which all error may be entirely obviated.

By placing the crystal so that the image of some distant object is brought to correspond with some other object by one of its surfaces, the position of that surface is determined with precision, and the second surface may he brought round to the same position with the utmost accuracy.

With this instrument the author has remarked an error in the supposed angle of the primitive crystal of carbonate of lime, which, instead of being 104° 28' 40", as it is now considered by writers on crystallography, appears to the author to be correctly 105°, as it was formerly measured by Huygens and by Sir Isaac Newton.