The object of this paper is to demonstrate the widespread occurrence of zoned olivines in igneous rocks, to show that the zoning is continuous, that the composition varies from a magnesium-rich centre to an iron-enriched margin, and that these findings have an important bearing on the question of petrogenesis.

I propose to express the composition of olivine by stating only the percentage of fayalite, as the rock-forming olivines belong mainly to the forsterite-fayalite series. There are, however, a number of names in use for the members of this series. Assuming that the two names forsterite and fayalite are insufficient for the needs of classification, it yet seems to me that all but four of the existing names are superfluous.

While crystal classes based upon symmetry furnish a clear-cut classification of crystals free from ambiguity and uncertainty, there are cogent reasons for recognizing the larger groups called systems.

It is important to recognize crystal systems, not only on account of physical properties, but also on account of the difficulty of assigning some crystals to their appropriate class. For example, a crystal consisting of a tetragonal prism and a pinakoid may belong to any one of seven classes. For the time being it is placed in the tetragonal system. When it is carefully investigated and its symmetry determined it may be assigned to the proper class.

Contrary to statements that have been made, we can have, and do have, a two-fold classification in geometrical crystallography—the thirty-two classes based upon symmetry, and the six systems based upon properties that certain classes have in common.

In the opinion of the writer the most satisfactory set of names for the crystal classes is the one based upon names of general forms. These names of Groth are comparatively simple, consistent, and lend themselves to international usage.

As is well known, quartz rarely shows well-defined cleavages. Kenngott has shown that quartz that has suddenly been cooled, after strong heating, has a tendency to break up in the directions of rhombohedron faces. Breithaupt (cf. Hintze, Min., 1904, vol. 1, p. 1273) mentions ‘cleavage following the primary rhombohedron and the prism, rarely well defined, often passing into conchoidal fracture’.

It may be useful to put on record an example of clear rock-crystal in which cleavage is developed in all three 60° directions of the prism faces. The specimen was originally in the Daimeries collection, in Brussels, having been bought years ago at Idar, near Oberstein on the Nahe, by A. Daimeries. I obtained it from Mr. A. G. F. Gregory, who had bought the Daimeries collection. The locality from which it came is, unfortunately, unknown.

Sartorite is one of four minerals once thought to be isomorphous. Of these only two are truly so—chalcostibite, copper antimony sulphide, CuSbS2; and emplectite, copper bismuth sulphide, CuBiS2—both orthorhombic minerals. The third, zinckenite, has been shown by two of the authors to be hexagonal and to approach more closely the formula Pb6Sb14S27 ═ 6PbS.7Sb2S3 than PbSb2S4 ═ PbS.Sb2S3. It seemed appropriate to continue a study of the series by an X-ray investigation of sartorite, a lead arsenic sulphide formerly supposed isomorphous with zinckenite. More than usual interest was attached to such a task, since discrepant interpretations of the symmetry and setting of sartorite had been obtained by well-known crystallographers.

Indeed, the difficulties of indexing different crystals of sartorite on the same lattice proved insuperable to G. F. H. Smith and R. H. Solly, so that they devised three incongruent lattices—one monoclinie and two anorthic. Also they used slightly different elements for each crystal. Their device, however, was not uniformly successful with certain crystals of sartorite previously measured by C. O. Trechmann. Two crystals in particular eluded their efforts. These, they decided, could not belong to the same species, but must be another mineral, perhaps a dimorphous form of sartorite, which they then named sartorite-α.

The recent finding of a stellerite-like mineral in California has led to this study of its relations to epidesmine and stilbite. The mineral was found by Mr. Allen Nichol, mineralogist of the Division of Highways, in the quarry of the Calavera Materials Company near Oceanside, San Diego County. Some specimens reached the writer through the State Division of Mines. The mineral occurs as sheaf-like clusters of small glistening white crystals in cracks and cavities of andesite. The largest crystals are nearly a quarter of an inch long, but tend to aggregate formation so that they do not lend themselves to accurate goniometric measurement. They are apparently orthorhombic, of tabular habit, showing the three pinakoids and small bright faces of a pyramid. The excellent pinakoid cleavage has a pearly lustre, and if this is taken as (010) the crystals, which show parallel extinction, have the same optical orientation as stellerite.

Some time ago, Mr. F. C. Partridge, Mineralogist to the Geological Survey of South Africa, gave me some crystals of sapphirinc which he thought might be of interest for goniometric examination. They come from the farm Blinkwater 1037, about 47 miles NNW. of Pietersburg and 45 miles WSW. of Louis Trichardt in northern Transvaal, and were collected by him in July 1930, after he had identified some specimens of sapphirine sent to him by Mr. A. Ruskovitch of Abrasive, northern Transvaal. They occur in a coarse granitic rock rich in brown mica, and are associated with small prismatic crystals of pink corundum.

The sapphirine crystals are almost invariably tabular or platy in habit, although occasional crystals possess a rhomb-shaped cross-section. The largest crystal measures 50×35×5 mm., while the thickest one is only 7 mm. in thickness. They are generally six-sided in outline with a strong suggestion of orthorhombic symmetry.