Presenting as it does the only example of a volcano in the phase of permanent moderate activity to be found in Europe, Stromboli must always have the strongest claims on the attention of geologists. Here may at all times be witnessed, in perfect security, those explosions produced by the disengagement of vapour in the midst of masses of liquefied rock, which, following one another at longer or shorter intervals, and taking place with greater or less violence, constitute a most striking feature in nearly all volcanic eruptions; and the causes, sequence and attendant phenomena of these outbursts can in the case of this volcano be most conveniently studied.

Tschermak examined two microscopic sections of the Ovifak rocks, and compared them with sections of the meteorites of Jonsac, Juvinas, Petersburg, and Stannern, which consist chiefly of augite and anorthite, with little or no nickel-iron; they form a class which G. Rose termed ‘eucritic.’ Both sections exhibit a crust, as meteorites possess; it is, however, so altered by oxidation, that it is not possible to determine whether it is the fused crust usually noticed on a meteorite. The crystals of felspar, which, according to Nauckhoff's analyses, must be regarded as anorthite, are fully developed; they penetrate the augite, iron, and magnetite, and must evidently have been formed before them. They are completely transparent, and have but few and large cavities, which are filled, partly with black granules, partly with a brown substance of irregular form; some traversing the length of the crystals are filled with a transparent glassy substance. The augite is of a light greenish-brown hue, traversed here and there by flaws; it fills gaps between the other constituents, as has been often observed in dolerites and diabases, and encloses individual black grains. In the section containing iron the colourless felspar encloses a black or brown substance running the length of the crystals, or dust-like fine black granules, or larger round transparent bodies of a violet colour, which may be the mineral Nauckhoff regards as spinel.

Much has been written on the relations of the Devonshire Trias as observed in the south-coast section. The subject itself is associated with many names of high scientific repute, so that, were the following epitome the result of partial examination, or in any way aided by preconceived notions arrived at from the perusal of the labours of those who have gone before, I should not feel justified in differing in many points from men infinitely my superiors in general geological information.

The publication in the Number of the Geological Magazine for November, 1874, of Mr. Goodchild's ingenious Drift theory leads me to make a few remarks on the subject.

In the paper on “Drift” in the Geological Magazine for November, 1874, Mr. Goodchild, in proposing his new theory, appears to invite discussion on it. I therefore beg to contribute my quota towards upholding the marine, as against the exclusively glacial, deposition of certain gravels belonging to the Drift.

The typical forms of Chœtetes petropolitanus, Pander, are free in habit, and are invariably furnished with a flattened or concave base, which is covered with a thin concentrically-wrinkled epitheca. The general shape of the corallum is typically a cone or hemisphere when full-grown, and a concavo-convex disc when young; but its adult form is liable to great variations; and even the young examples are not constant, being sometimes plano-convex, or slightly bi-convex. The calices are exclusively carried upon the upper surface of the mass, and are thin-walled, polygonal in shape, and destitute of intermediate tubuli. Very often the surface exhibits minute tubercles, carrying corallites of a slightly larger size than the average; but this phenomenon does not appear to be constant. Young examples may have a diameter at their base of four or five lines, and a height of three lines; adult specimens may have a long diameter of four inches or more at the base, and a height of an inch and a half; but the majority of individuals exhibit dimensions intermediate between these measurements.