During the investigation of a zeolite it became necessary to separate small quantities of zeolitic material from a mixture of calcite and other heavier minerals; this could be done most easily by using a liquid of such density that the zeolite just floated, while all the other minerals sank to the bottom. A solution of potassium-mercuric iodide could not be used, as this reacts chemically with zeolites and alters their composition ; acetylene tetrabromide diluted with benzene was therefore chosen, and in consequence the apparatus had to be made entirely of glass. Certain dittlculties were found in using the ordinary form of glass separating apparatus, and to overcome these the form here described was devised.

A preliminary notice of this new mineral, named in honour of Dr. Arthur Hutchinson, Demonstrator of Mineralogy in the University of Cambridge, was read before the Cambridge Philosophical Society in October, 1903. The material was first found, in the white crystalline dolomite of the Lengenbach quarry, during my visit to the Binnenthal in the summer of 1903. More material was received in November of the same year, and a further supply was obtained in 1904.

The first form of refractometer, in which a reference-scale was substituted for the rotational parts and graduated circles of the more complex instruments, was designed in 1885 by Professor E. Bertrand who at the same time was the first to suggest the employment of a hemisphere of dense glass in place of the prism and cylinder previously used. The original instrument and others on the same principle which have hitherto been constructed do not give satisfactory results, because no attempt has been made to compensate for the curvature of the focal surface of the hemisphere, and the determinations obtained are, therefore, far from trustworthy. In the refractometer described in this paper a corrective lens is placed in the path of the rays emerging from the hemisphere, and the field thus produced is rendered very nearly plane.

The determination of the amount and sign of the relative retardation of the wave-fronts of the two rays, into which a plane-polarized ray of light is resolved in passing through an anisotropic crystal, furnishes us with the readiest means of identifying transparent minerals occurring in thin sections or as minute fragments without definite crystalline form. In sections which are of practically the same thickness the individual crystals of each mineral show a certain range of colours resulting from the interference of the rays after resolution parallel to the principal plane of the analyser, and these colours are usually sufficient, with the help of other characters, for the identification of the mineral species.

Since the mineral gyrolite was described and analysed by Anderson from Storr in Skye in 1851, it has been recorded from a considerable number of other places in that island. The writer observed it at the mouth of the burn called Allt Mòr, between Loch Brittle and Loch Eynort, and has now to record it from Rudha nan Clach at the mouth of Loch Bracadale, which brings the number of localities at present known in Skye up to fourteen. In fact, the mineral may be said to occur all along the south-western shore, all along the eastern escarpment, and at the more isolated localities of Lyndale and Portree. Similarly, it occurs in the other trap islands—the Treshnish group, Mull, Muck, Eigg, Canna, and Sanday, upon the last of which it has been observed by the writer at Stac nan Faoilean.

The references to the occurrence of the titanium dioxide minerals as rock constituents have been so recently summarized in the ‘Mineralogical Magazine, that it is unnecessary to repeat them, except to refer again to Mr. J. J. H. Teall's record of anatase in the Cleveland ironstone.

Wishing to procure some of these crystals, a pound of ore from the Main Seam at Upleatham mines was treated by the usual processes of solution, washing by decantation, and separation by heavy solution. The residue obtained was equal to 0.16 per cent. of the material taken, and consisted mainly of anatase, mostly in sharply developed pyramidal crystals, with some pyrites in nodular concretions, possibly in replacement of some of the oolitic grains, together with other minerals of derived origin, among which futile and tourmaline were the most prominent.

§ 1. It is proposed in this paper to describe some simple graphical constructions of theoretical or practical interest. They may be easily performed with ruler and compasses, though the stereographic and gnomonic nets will often save labour. To economize space the proofs of the constructions are left to the reader.

§ 2. First take the case of the gnomonie projection. Given the projections A, B, C, Q (fig. 1) of the poles of the faces (100), (010), (001) (111); find the projection of the pole of any other face (hkl).

Let the crystallographic axes OA, OB, OC of a crystal, whose axial ratios are a : b : c, meet a plane p in A, B, C. Let a plane through O, parallel to a crystal-face whose indices are h, k, l, meet p in the line f (fig. 1) ; and let the line OF, perpendicular to the face, meet p at the point F. Let the perpendicular from O on p meet it in V, and let OV = m.

The mineral to be described was found by Mr. C. T. Clough, M.A., of the Geological Survey, near the junction of the two large streams in Coire Dhorrcail, near Barrisdale on the south side of Loch Hourn, Inverness-shire.

Mr. Clough furnishes me with the following description of the mode of occurrence of the mineral :—

‘The crystals were obtained from a band, which, although its sides are not well exposed, seems to form a nearly horizontal vein-like mass. This band crosses both the bedding and the foliation planes of the muscovite-biotite-gneiss in which it occurs, and which occupies a large area near Barrisdale and Ladhar Bheinn.