In the previous paper I ventured to trace the history of the Baltic back to the time when the latest of its raised beaches were laid down, and to show that it was then considerably larger in size, and that its waters were more salt than they are now, although they were even then brackish (see Map I, Pl. XIX).

This conclusion was derived mainly from an examination of the molluscan remains in the more recent shell-beds. It is confirmed by other evidence; thus, Munthe mentions the occurrence in the so-called Litorina beds of three species of Rhizopods, whose present and former distribution in the Baltic is shown in the following table:

What represents the Carboniferous Limestone in Devonshire, is a question that has often been asked, but has never yet been answered in a satisfactory manner. The question has recently been narrowed by the identification of the Coddon Hill Beds and the Posidonomya limestones with the shales, cherts, and limestones which overlie the mass of the Carboniferous Limestone in South Wales, and with the Pendleside Beds which occupy a similar position in Staffordshire, Derbyshire, and South Yorkshire. Dr. W. Hind considers these beds to be more intimately connected with the Upper than with the Lower Carboniferous Series, and in any case the homotaxial equivalent of the main mass of the Carboniferous Limestone must be sought for below them.

Nearly forty years ago Mr. R. D. Darbishire obtained from the Glacial gravel of Bowdon, Cheshire, a facetted pebble, which I afterwards had the privilege of describing (Proc. Geol. Assoc., xvi, pp. 396–420, pl. xi, June, 1900). The conclusion then reached was that the pebble bore witness to the action of blown sand at a period during or not long before the deposition of that particular bed of Glacial Drift. Since, however, the pebble was in other respects similar to pebbles common in the Bunter beds, and since some had supposed the existence of desert conditions in those Triassic times, there remained the possibility, and to some minds the probability that the pebble in question had received its facets during the Triassic epoch. It may, therefore, be useful to draw attention to a distinctly facetted pebble from the Boulder Clay within twenty -five miles of Bowdon. This was found by Mr. W. D. Brown, of Burscough (north-west of Ormskirk, Lancashire), at about three feet below the surface in the Burscough Brick Company's clay pit. The pebble is composed of a greyish micaceous sandstone, weathering yellow, and too coarse and loose in texture to be capable of taking a polish. The facets on it, however, are clearly cut, and do not appear to have undergone much, if any, subsequent abrasion. Its greatest length is 37.5 mm.; greatest width, 29.5 mm.; greatest thickness, 7.9 mm. What was probably its uppermost surface, during the latest period of its exposure to wind action, shows a flat central area, of irregularly pentagonal outline, surrounded by five facets, of which one is slightly concave, one rather irregular, two slightly convex, and the last markedly convex and probably retaining the original curved surface of the pebble.

What are popularly termed the Banket Beds of Johannesburg have for years past been justly celebrated for their extraordinarily large and steady gold production. Their literature has already assumed formidable dimensions, so it is unnecessary for my present purpose to give a detailed account of the features they present. They form part of what is generally known as the Witwatersrand Series, which consists of highly altered sandstones and conglomerates almost certainly of Archæan age. They unconformably overlie the banded rocks known as the Hospital Hill Beds, and are overlain by the Dolomite, often termed the ‘Olifants Klip’ or ‘Elephant Rock.’

The object of the present paper is to provide a short account of a group of rocks to which the general name of pyroxenite may be given, inasmuch as they are essentially non-felspathic plutonic rocks, free from olivine, and of igneous origin, and with a nearly colourless monoclinic pyroxene as the chief or only constituent of many of the most conspicuous varieties. The exposures are widely distributed in the Central, Ova, and Sabaragamuwa provinces, and probably throughout Ceylon, but are individually of very small extent. The rocks are composed of the minerals diopside, amphibole, phlogopite (or biotite), scapolite, pyrite, sphene, and sometimes felspar, calcite, and spinel; the dominant mineral is the pale green or grey granular diopside, colourless in thin sections: the rocks are often entirely, or almost entirely, composed of this mineral; next in importance are green amphibole (dark in hand-specimens, but very pale in thin sections), and a pale golden-brown phlogopite, which sometimes forms a large proportion of the rock, and in extreme cases the whole bulk of sills or dykes several feet in thickness.

In 1895 the author visited the Prince Albert Goldfields, and later the Millwood Goldfields in Knysna, and has had the subject constantly before him till quite recently, when, in reviewing the whole question in connection with the geology of the region, which he has now worked out under the auspices of the Geological Commission of the Colony of the Cape of Good Hope, he was able to come to some definite conclusions. The gradual building up of convictions on a subject like this, as facts became clearer with increased knowledge derived from the reports of prospectors, the progress of the Survey, and the access to literature bearing on kindred questions, led to the acceptance of certain general principles, which the author has thought worth while to state in definite form for criticism. The peculiar points in the occurrence of the precious metal in these localities cause them to rank high as fields for research, if not for general enrichment of the inhabitants of the Cape Colony.