Some years ago I was asked to look through a collection of geological specimens in the Raffles Museum, Singapore, and found in a drawer two exceptionally large obsidianites. They were not labelled, and nothing could be ascertained about their history, but an assistant in the Museum said he thought they might have come from Kelantan, a State on the east coast of the peninsula. The weights of the two obsidianites were 464 and 316.4 grams. The former and larger of the two was entrusted to a local firm to cut in half, with the result seen in Pl. VII, Fig. 1. The photograph, nevertheless, shows that there is a group of vesicles in the centre. The photographs, which are natural size, also show that there is nothing unusual about the surface of these specimens (Pl. VII, Fig. 2).

The structures described below occur in rocks of a rather monotonous and generally uninteresting type; namely in sandstones, interstratified with masses of shale and mudstone, exposed near the centre of the Berwyn Anticline in parts of Denbighshire. The necessity, however, of examining every exposure, incumbent upon the Surveyor, has led me to pay attention to detail, and especially to certain phenomena that would appear to be of world-wide occurrence in sediments of all ages.

In a previous paper on the Permian Formation in the Alps of Piémont, Dauphiné, and Savoy, I referred incidentally to the large masses of pietre verdi or greenstones which constitute perhaps the most striking geological feature of the extensive areas covered by the crystalline rocks of the Piémontese Alps in a crescent-shaped curve about 200 miles in length from the Maritime range to Monte Viso, Grand Paradiso, and Monte Rosa. In the present paper I propose to deal more fully, although necessarily within narrow limits, with these pietre verdi which, owing alike to their extraordinary development, variety, and complexity, to their intimate association with each other and with the crystalline sedimentary rocks, and to their intricate composition and origin, have for the last fifty years presented most interesting problems and passed through many remarkable phases of interpretation. As a necessary preliminary to a description of the different areas, it will be convenient to briefly consider the most recent classification of the crystalline formations of the Piémontese Alps generally, and of the pietre verdi rocks in particular.

Although detrital mineral work is as yet in its infancy, sufficient has been accomplished to show that we may look to it with success for indications of changes of drainage direction, evidences of denudation by reversal of the order of respective mineral assemblages from a sequence of rocks, and generally for information regarding details of palæogeography. Professor A. de Lapparent referred to Professor L. Cayeux's work as proving the proximity of land, composed of primary rocks, to Lille in Landénian times. Dr. H. H. Thomas was able to demonstrate the change in source, and therefore in direction of drainage, of the river-borne heavy minerals in the Bunter sandstones of South Devon, the occurrence of garnets and staurolite being especially significant. Dr. T. O. Bosworth, in some preliminary work upon the detrital minerals of the Carboniferous Sandstone of the Midland Valley of Scotland, was led to the conclusion, partly by the respective presence and absence of garnets, that the beds could be divided into a series of great lenticular masses of sediment introduced from directions varying from north and north-west to north-east, east, and south. Mr. W. R. Smellie has discussed in rather more detail the origin of the minerals in the Upper Red Barren Measures of the Glasgow Basin, the drainage having been from the west or north-west.

We have now finished the survey of the characters of the head-shield of Trinucleus sufficiently to enable us to discuss the vexed questions of the affinities, systematic position, and evolution of the genus; and from the foregoing details of the various species we can perceive that there are many difficult problems presented when we attempt to correlate its structure with that of other trilobites. Whether the peculiar characters of the genus are the result of degeneration or of the persistent retention of early phylogenetic stages or of a reversion to such stages is a matter of dispute.