Experiments made by Professor Miers and Miss F. Isaac have shown how, by measurements of the refractive index, it is possible to determine for a supersaturated solution the temperature at which it passes from the metastable condition, in which it can only crystallize by inoculation with a crystal of the salt (or of a substance isomorphous with it), to the labile condition, in which it may (and often does) crystallize spontaneously. In a diagram in which concentrations are taken as ordinates and temperatures a abscissae the solubility curve represents the limit between the unsaturated and the metastable state; and the limit between the metastable and the labile state is represented by a similar curve, to which they have given the name ‘Supersolubility curve’.

As has been mentioned in the preceding communication, a supersaturated solution, which is cooling or evaporating, passes successively into the metastable condition, in which an introduced crystal can grow in it, and then into the labile condition, in which it may crystallize spontaneously. The transition is indicated by a sudden change in the refractive index of the solution. It seems possible that in these two distinct states of the solution there may be a difference in the mode of growth of the crystals. In the following experiments an attempt was made to find evidence of such a difference by the examination of thin drops of solution crystallizing under the microscope: and potash-alum was found to be a convenient substance for the purpose.

Flexible mineral fibres appear to have been well known to the Greeks, who spun them into thread, which they wove into cloth or employed as wicks for lamps.

Pausanias, who lived in the second century of the present era, tells us that Kallimachos made for the statue of Minerva in the Acropolis at Athens a lamp of gold, the wick of which was of ‘karpasiau flax’, which he states is the only flax inconsumable by fire. Some have interpreted the words ‘karpasian flax’,——as meaning mineral flax from the Karpasos, the north-eastern promontory of Cyprus, but apart from the fact that there is no record of the occurrence of asbestiform minerals in that locality, it seems clear that the word is derived from (feminine in the singular, but neuter in the plural ), which properly signifies cotton, being identical with the Sanskrit , though it has been used for a fine variety of flax produced in Spain, as well as for other European plants.

The gnomonic projection in which the faces of crystals are represented by the points where their normals through a fixed point–the centre of projection–meet a plane–the plane of projection–is already widely used by crystallographers. I propose in the following pages to show the advantages of the simultaneous use of two such planes of projection.

It is in most cases convenient for the two planes to be at right angles, and for the centre of projection to be at an equal (unit) distance from each. Circumstances, however, in some cases render it desirable to vary these conditions. When the planes are at right angles, one may conveniently be described as the horizontal plane, and the other as the vertical plane.

It will in the first place be assumed for the sake of simplicity that the crystal to be investigated is immersed in a liquid having a refractive index as nearly as possible equal to the intermediate refractive index (β), so that light travelling along the optic axis t is not appreciably refracted on entering or leaving the crystal.

The observation is made in parallel ight between crossed nicols, which are rotated together till a position of extinction is reached, when the directions of vibration of the nicols wilI be parallel to those in the crystal. If the latter be now rotated on an axis parallel to one of these directions, it will in general show light throughout the revolution, except in the original position and that opposite to it. Occasionally, however, it will remain dark throughout; in that case the axis of rotation is the bisectrix of an optic axial angle.

In connexion with the systematic examination of the minerals of Ceylon, which is being conducted in the Scientific and Technical Department of the Imperial Institute, the authors have recently examined a large quantity of coarse gem-gravel refuse obtained from the Balangeda and Rakwana districts of Ceylon.

The material varies somewhat in character, but zircon generally pre-dominates, while tourmaline, corundum, and pleonaste are fairly abundant. Among other minerals present in noticeable quantities are geikielite, picroilmenite, ordinary ilmenite, rutile, and fergusonite, all of which are generally in the form of rounded or sub-angular pebbles.

The mineral dundasite was first described by W. F. Petterd in 1893, a ‘Catalogue of Minerals known to occur in Tasmania’. It was found in small, silky, milk-white, spherical aggregates incrusting a ‘ferro-manganese gossan’ at the Adelaide Proprietary mine, Dundas. In this catalogue, and in a subsequent one dated 1896, it was stated, as the result of qualitative tests, to be a hydrated carbono-phosphate of lead and aluminium ; but a quantitative analysis (see below under II), made later by S. Pascoe, showed that the mineral is a hydrated carbonate of lead and aluminium, and that the traces of phosphoric acid were due to a thin coating of pyromorphite. In this analysis the water and carbonic acid were determined together, by loss on ignition, and consequently hitherto no chemical formula has been attributed to the mineral.

About two years ago there was brought to the British Museum a specimen covered with numerous mall cubes—so the crystals appeared to be—of the bluish-green colour suggestive of a mineral containing copper. Judged by their form the crystals might be beleito, but a qualitative analysis by Dr. Prior, while revealing the presence of chlorine and copper, showed no trace of lead. A crystal was then removed from the specimen and measured on a goniometer, and the discovery was at once made that these apparent cubes are iu reality rhombo-hedra with an angle of nearly 88° between adjacent faces. Since no previously known oxychloride of copper crystallizes in this form, these crystals were evidently worthy of further investigation.

A small collection of minerals, together with other natural history specimens, has recently been acquired by the British Museum from the Rev. A. North Wood, a missionary in the Usagara (Ussagara) province of German East Africa. Mr. Wood's station is situated at Itumba, and the specimens had been brought o him by the natives from the surrounding district, mainly from places between Mamboya and Mpapwa, which lie on the caravan route from the coast opposite Zanzibar westwards into the interior to Lake Tanganyika.

The rocks of the district are biotite-gneiss and hornblende-gneiss, the latter with small garnets.

Some very brilliant and well-modified crystals of ilmenlte, recently found in the neighbourhood of the Ofenhorn, are of interest in that they exhibit a well-marked parallel hemihedrism and show the presence of a number of new forms. Although constantly sought for, it is more than fifteen years since crystals of this mineral were last found in the Binnenthal. The crystals described by H. Bücking in 1877 appear to differ in habit from those of the new find.

A determination of the refractive power, which need be accurate to the second place of decimals only, is of considerable service for the discrimination of transparent substances, such as, in particular, faceted gem-stones. If the substance be so highly refractive that a total-reflectometer be not available, the method of minimum deviation is the simplest hat can be used.