The mineral here described was at first considered to be an iron-rich variety of turquoise and was described by myself as such when this paper was read before the Society in 1937. At that time only a partial analysis had been made; since then Dr. J. A. Smythe of King's College, Newcastle-upon-Tyne, has most kindly undertaken at my instigation careful analyses of the mineral from both of its localities, Bunny mine, St. Austell, and Castle-an-Dinas wolfram mine, St. Columb Major. As a result of these two analyses, which are in close agreement, it is evident that the mineral is a new one, intermediate between turquoise and the two minerals chalcosiderite and andrewsite, and forming a middle member of what is probably an isomorphous group.

Many examples of the hydrothermal synthesis of minerals are scattered throughout mineralogical literature. The present paper has been written to bring to the attention of mineralogists the fact that such hydrothermal syntheses are occurring continually in industrial boilers, particularly when control of water softening procedure is not completely efficient.

The method described is one that can be used to measure the refractive indices of minerals when no refractometer is available. The readings are consistent to ±0·002 and are in reasonable agreement with those obtained using a refractometer.

The principle of the method is to mix two liquids with different refractive indices. One of these is a typical immersion liquid with a very low volatility. It must have a refractive index greater than that of the mineral under investigation. The other, more volatile liquid has a refractive index less than that of the mineral. The two liquids selected must be miscible in all proportions, but must not react together. The mineral is completely immersed in a mixture of the two liquids.

A laboratory investigation in connexion with some blast-furnace slags, led to the preparation of the five synthetic melilites for which X-ray data are provided. The five compounds represent gehlenite, åkermanite and three members of the intermediate series of solid solutions, corresponding to 25, 50, and 75 % of åkermanite.

The binary system gehlenite–åkermanite was studied by Ferguson and Buddington, who established relationships between refractive indices, density, and composition and determined solidus and liquidus curves. The crystal structure of the melilite group of compounds was investigated by Warren, who showed that the structure was based on a tetragonal lattice.

The andesine-dolerite of Squilver Hill, situated nearly 3 miles due north of Bishops Castle, has been described in another place. This somewhat coarse-textured rock, with a distinctive greenish-blue colour, contains up to 20 per cent. of a fresh monoclinic pyroxene as crystals of irregular shape, many of which are about a millimetre in length. The rest of the rock consists of saussuritized plagioclase, some interstitial chlorite and prehnite, and accessory apatite, sphene, magnetite, and ilmenite; the unaltered nature of the pyroxene is noteworthy.

Phase-contrast methods in microscopy appear to have been developed mainly with the intention of applying them to biological subjects, and their application to the examination of mineral substances has received only slight attention. Thin sections of rocks viewed under a phase-contrast microscope have a fascinatingly strange appearance and it does not take long to realize that some features are shown up more clearly and others less clearly than when viewed under an ordinary or a polarizing microscope. A considerable amount of work may be necessary before it is possible to assess the value of the new instrument for petrological purposes or to interpret all the phenomena observed by its aid. Nevertheless, it seems fitting to set out the following short account as a contribution towards this knowledge.