Very little has been written about the Coelacanths from Madagascar, although those from similar formations in East Greenland and Spitsbergen are about the best known members of this group. Smith Woodward (1910) described and figured a single specimen as Coelacanthus madagascariensis, distinguished from other members of the genus by the ornament of closely set tubercles on the operculum. Further specimens were figured by Priem (1924) and attributed to C. madagascariensis, but no attempt at any detailed description was made. Dr. E. I. White, of the British Museum (Natural History), very kindly placed at my disposal a collection of these fish, which he had obtained himself. The material examined during this investigation consisted of these specimens, twenty-eight in number, and the original specimen described by Smith Woodward. The specimens were nearly all external moulds (although in some a certain amount of bone still remained) in ferruginous clay nodules. Very little preparation was necessary except washing. This treatment produced wonderful results in some cases and almost perfect impressions of the head were obtained. So perfect were these casts in many cases that the pores of the lateral line canal were left as small hummocks making parts of this system easy to identify.

The correlation of isolated outcrops of igneous rocks where two or more similar intrusions are exposed is a difficult problem which is not always solved by thin sections or field studies. Such a problem was encountered in mapping the closely related Pikes Peak and Silver Plume granites of pre-Cambrian age in the Sawatch Range of central Colorado (Fig. 1). A comparison of the heavy minerals of the isolated outcrops with those of known granites was undertaken; and for this purpose large samples, suitable for crushing and heavy mineral analysis, were collected from various points within the areas of each batholith, and from the small outcrops whose age was in question. It was hoped that sufficient similarities in the heavy mineral assemblages might be established to be of value in making correlations. Furthermore, as work on the heavy minerals in igneous rocks is still in the experimental stage, a series of analyses from various parts of a given batholith should throw some light on the question of whether heavy minerals may be distinct and constant enough to be characteristic and so give a reliable means of correlation of isolated exposures.

CYATHAXONIA siluriensis M’Coy is a small and rather rare rugose coral, which persists from the Woolhope Limestone to the Upper Ludlow. It is here placed as the genotype of Syringaxon Lindström, and an account is given of its structure and development.

Recently the author has been investigating the sequence of floras and faunas in the Warwickshire Coalfield. In the course of this work particular attention has been given to the flora of the barren Upper Coal Measures, which, according to Eastwood (1923, p. 42), includes the Keele Series and at least 1,700 feet of the Corley or Enville Series. Vernon, on the other hand, placed the latter in the Permian (1912, pp. 590, 603), and he noted that the Corley Series rests unconformably on the underlying Upper Carboniferous strata. Shotton, however, stated (1929, p. 169) that in this area he could see “no evidence for any Permian System (in the sense of System being separated from the Carboniferous by an unreprented time-interval)”.

The two chief difficulties in taking serial sections are to ensure that the successive sections are all parallel to each other and to obtain reliable drawings of them.

The series of estuarine deposits which occurs at Kirmington, North Lincolnshire, situated between Brown Boulder Clay and Purple Boulder Clay, has long been recognized as of considerable importance owing to its inter-glacial position. The best description of the complete sequence of beds overlying the Chalk at this locality has been afforded by a Committee of the British Association (1).

A small collection of Triassic fossils from three localities and two horizons has been sent me by Dr. Euzebio de Oliveira for determination. Some of them (Nos. 1–10, 12–18) come from Rio Claro, Canoinhas, Sta. Catharina, and occur in the same hard siliceous oolitic rock as those which were first described from this locality. One specimen (No. 11) in a similar matrix comes from Colonia Vieira, Canoinhas, and all these specimens were collected by Aristomenas Duarte. They belong to the fauna which was described by the author from Rio Claro and other localities in 1929 and 1932, and a long list of fossils was then given. But we have now to add three new species.

In 1836 Rose (Phil. Mag., viii, p. 30) described a series of marine deposits in the Nar Valley, associated with the brick clay of that neighbourhood. Later work revealed that these beds were widely distributed to the south-east and east of King’s Lynn, and were to be correlated with very similar gravels at March and Whittlesey. Further correlation outside this area is uncertain, but following Boswell’s scheme (1932, pp. 72, 75) the March-Nar marine episode may be classed with the well-known estuarine sections at Kirmington, Lines.: moreover, he refers all gravels containing Corbicula (including the Barrington gravels) to this period. The same author regards the uppermost boulder clay at Kirmington as of Upper Purple age, and “apparently not the equivalent…of the Brown boulder clay of Hunstanton”.

In many coalfields in the zone of Anthracomya pulchra and Carbonicola similis there is a marine band, viz. Mansfield, Yorkshire; Dukinfield, S.E. Lancashire; The Gin Mine, North Staffordshire; Rimbert, North of France; but, apart from a doubtful Lingula from the roof of the Brassy coal at Bersham Colliery, no marine fossils have previously been recorded from undisputed Coal Measures of the Denbighshire Coalfield.

Between the Cruachan Granite complex and the Glen Strae fault, to the south-east, and the Pass of Brander fault, to the south-west, there is a strip of Dalradian rocks (Fig. 1). These all lie within 1 in. Sheet 45 (Scot.) of the Geological Survey and are described in the Oban and Dalmally Memoir (Hill, 1908, pp. 19–28; Kynaston, 1908, pp. 38–40). Bailey has later distinguished the three great structural divisions into which he has divided the South-West Highlands. These are, in ascending structural succession: the Ballapel Foundation, represented south-east of the granite by Eilde Flags, Glencoe Quartzite, Leven Schists, and Ballachulish Limestone, and south-west of the same by Leven Schists; the Iltay Nappe, represented in both these positions by Islay Quartzite and Easdale Slates; and the Loch Awe Nappe, represented south of the granite by Ardrishaig Phyllites (Bailey, 1922).

The general results which have accrued from this investigation may be summarized as follows: the shaly type of Liassic sedimentation with much pyrite, indicating a quiet sea with abundant sulphur compounds, gradually gave place to a regular and at first fairly steady shallowing of the water with, somewhat later, occasional incursions of strong currents bringing pebbles and broken fossils from an area undergoing active erosion. At the same time the amount of iron introduced, both in solution and in the form of derived ooliths, increased steadily, culminating in the siliceous oolite of the Dogger, which is essentially similar to the rocks composing this formation to the west and north-west.

It has been suggested that the grain-size of coarse basic plutonic rocks may be reduced in one or all of three ways:— (1) If the temperature of the invading magma be sufficient, the reduction may be effected by the formation of a granoblastic structure. It is pointed out that recrystallized inclusions are found more often in basic plutonic masses, where the earlier crystalloblastic structures are not obliterated by superimposed hybridization. In 1933 (p. 156) the writer suggested that relict felspars with granular inclusions (or clouding) may be useful criteria of a recrystallization antedating hybridization. (2) The grain-size may be reduced by the development of highly poikilitic crystals of hornblende and/or biotite, and by the final disruption of these. Such crystals may be formed as poikiloblasts in the initial recrystallization; but more often they may be attributed to hybridization, and in this process hornblende plays an important rôle. (3) The grain-size may be reduced by the formation of granular masses of pyroxene or by criss-cross flakes of biotite during hybridization. Thomas and Campbell Smith (1932) and Nockolds (1932) have discussed the development of large felspars within the fine grained xenolith and its final incorporation in the acid magma, but it is not the purpose of the present paper to deal with this stage. It is hoped that this discussion may have indicated a means by which coarse basic plutonic rocks may become fine grained xenoliths, and it is finally urged that it is not always necessary to postulate a fine grained origin for these inclusions.

The genus Tricycloseris was erected by R. F. Tomes in 1878 (p. 190) for a single specimen of a compound coral obtained from a fallen block on the shore near Charmouth, Dorset. The associated fauna led him to postulate a Middle Lias age for the bed from which the block was derived. A recent examination of the specimen shows that Tomes’s diagnosis was unfortunately based on a complete misconception of the coral—what he termed the “upper surface” is actually the under surface, and what he referred to as the “under surface” is, in reality, the calicular surface.

From time to time measurements of the rate of sedimentation in salt marsh areas have been made, but few detailed figures are available. The fullest account is that by F. J. Richards (1) for the Dovey Marshes. Others have been made at Blakeney by F. W. Oliver (2) and, recently, measurements have been made in Denmark by Dr. Niels Nielsen, of Copenhagen, and at other continental stations.

During the last twenty years the Pleistocene of Central Europe was investigated by a great number of geologists. As a result it is now possible to establish a detailed stratigraphy which proves to be generally applicable. When employed in conjunction with certain astronomical results it enables horizons, fossils, and implements of many localitiesto be dated in thousands of years.

A comparison of the chemical features of Scottish magmas, has been greatly simplified through the appearance of a recent memoir, compiled from the records of the Geological Survey (Guppy, 1931). This memoir gives the composition, age, locality, and microscopic petrography of a great number of Scottish rocks. The present note deals with three Scottish magmas: Devonian (Lower Old Red Sandstone), Carboniferous and Tertiary. It excludes certain analyses of the Survey list, owing to decomposition of the analysed material. It includes a few from outside sources, to augment the data concerning the Devonian magma (F. Walker, 1924, 1925, 1927, 1928; and J. G. C. Anderson, 1935).

Consideration of the continental areas over which the sea spread at various times during the course of geological history leads to the impression that perhaps every part of the land surface of the earth was, at one time or another, raised from a position below sea-level. The present land surface seems to have grown around ancient nuclei owing to the compression of the granitic, sedimentary, and other rocks of which the continental layers are composed. At certain times the compressive forces acted so strongly upon these rocks that the average height of the land reached maximum values, which might have been as great as 3,000 feet. Then denudation, and perhaps other processes, coming into operation and persisting through ages of comparative quiescence, lowered the surface, sometimes to minimum average levels, which might have been as low as 500 feet. There have been a few comparatively short periods of exceptionally high relief, separated by longer periods of low relief. Assuming that the land was raised by horizontal compression, the diminution in surface area of the earth necessary to produce a change in level of 2,500 feet would be very considerable. Instead of considering the change from a minimum average height of 500 feet to a maximum of 3,000 feet, it may be more convenient in the first place to consider the elevation of, say, the continental layers of Dr. H. Jeffreys from sea-level to the average height of the present land surface, that is about 2,500 feet, or 762 metres. It will be supposed, then, that the continental layers, as they now exist, were developed from layers consisting of 1 km. of sediments of specific gravity 2·4, 10 km. of granite of specific gravity 2·6, and 20 km. of tachylyte of specific gravity 2·9, the whole resting on dunite of specific gravity 3·3.