The specimen represented in Fig. 1 was found by Mr. J. A. Whaits at Prince Albert in shale lying between the Witteberg Series and the Dwyka conglomerate: it consists of a single whorl of leaves of which the longest has a length of 4–5 cm. and a breadth of 3 mm. The linear segments converge towards a narrow axis about 5mm. in diameter; they show faint indications of a midrib and have acuminate apices.

In spite of the fact that Fig. 1 represents the only example so far discovered I propose, for the sake of convenience, to describe it under a new name, thus placing on record Mr. Whaits's contribution to our knowledge of a flora the discovery of additional records of which would be of exceptional interest. It must, however, be admitted that the specimen for which the generictitle Phyllotheca is chosen hardly affords satisfactory evidence as to its affinity.

The material on which this species is founded was obtained from the Uitenhage Series at Herbertsdale in Cape Colony and presented to the South African Museum by Mr. Kolbe. It consists of several pieces of what appears to have been a large stem rèaching a length of about 90 cm. The outline drawing reproduced in Fig. 2 from a photograph sent to me from the Museum by Miss Wilman shows the curious shape of the fossil, tapering to a blunt point at one end and flatter and broader at the other; Figs. 3 and 4 illustrate the surface features as seen on portions of the weathered stem.

Round the south-western corner of Cape Colony there is a belt of mountains which exhibit a number of exceedingly interesting features. The ranges meet almost in a right angle, the one set trending north and the other east, from a point near Ceres village; this is what Suess calls a ‘Schaarung,’ and I believe nowhere else in the world does this structure show itself in such a simple manner. Each range is composed essentially of an S-shaped bend, the syncline on the coast side and the anticline on the inner side, while subsidiary folds are added in various places, without, however, obscuring the general nature of the mountain structure. The central axes of the ranges are composed ofa coarse grey false-bedded sandstone, the Table Mountain Sandstone resting on the outer side on Pal-Afric beds, slates, phyllites, and intrusive granite, and dipping under later and conformable beds on the inner sides. The age of the Table Mountain Sandstone is uncertain, as, with the exception of indeterminable bivalves found by Griesbach in Natal and possibly a Patella, no fossils have been recorded from the series; the overlying beds, however, the Bokkeveld Beds, contain a rich fauna belonging to the American type of Devonian species. On top of the Devonian conies another series of sandstones, the Witteberg Beds, strongly differentiated from the older sandstone series by their yellow and red tints, and the amount of shaly matter intercalated between the sandstone banks.The Witteberg Beds, although strictly conformable to the Bokkeveld Beds, contain a flora which is referable to the Lower Carboniferous, the conformity being supposed to be the result of the ocean floor, during the period occupied elsewhere with the deposition of the Middle and Upper Devonian, having sunk below the area, or rather out of reach of deposition of land detritus.

My attention was first drawn to this subject by the study of a remarkable crystal breccia, occurring at Kakanui, New Zealand, of which a description is given below. The only similar phenomena I could read of in the Colonial libraries at my disposal were those of the diamond pipes at Kimberley, described by Professors Carvill-Lewis, T. G. Bonney, and Miss Kaisin in this Magazine.2 Having since had the opportunity, thanks to the hospitality of Professor Lacroix, of studying in his laboratory at Paris, I found that the proper discussion of this subject comes under the study of the inclusions of volcanic rocks, of which he has given so admirable an account in his book “Les Enclaves des Roches volcaniques.”3 At his recommendation I collected some inclusions from the well-known dolerite of Portrush, and studied them under his guidance. The present paper consists of a discussion of rocks from these two localities, with descriptive notes on a few other inclusions that I have in my collection.

Two things stand out clearly from a cursoiy examination of this book, the frequency, almost the universality, of inclusions in igneous rocks, and the importance of the aid which they afford to the student of volcanic phenomena. Under these circumstances, it is remarkable how long this branch of study has escaped attention from. British geologists. Lacroix's useful nomenclature has not been adopted in our literature, although Phillips, Teall, Sollas, Harker, Judd,1 and others have paid considerable attention to some varieties of inclusion, and Harker has established a slightly different classificationof these phenomena.

On previous occasion (Geol. Mag., Dec. V, Vol. IV, 1907, pp. 17–20) an account was given by the author of lornagnt Strand in the estuary of the River Suir; and the nature anddevelopment of the drift deposits further up the river may now be described.

On the north side of Knockavelish Head (which bounds Fornaght Strand on the north), the coastfrom Ballyglan is bordered by low sand dunes for a distance of about a mile and a half, and thereis no cliff or exposure of any solid rock, the old sea margin being apparently situated now some way inland behind an area of more or less marshy laud at the back of the dunes. Traces of the former sea-cliff can, however, be recognised here and there by a sudden slight rise in the ground. A wide flat expanse of sand and mud, known as Woodstown Strand, is uncovered at low tide along this stretch of coast, extending out fully a mile from high-water mark, but so far no trace of the submerged forest of Fornaght has been here discovered.

While spending a few days recently examining the sections on the North Devon coast, with a view to obtaining a clearer conception of the tectonic structure of that region, I was deeply impressed by the great number of minor overfolds and with the great uncertainty in the relations of the various beds thereby brought about. In many cases where the dip appears to be constantly to the south, through a considerable thickness of strata, closer examination shows repeated isoclinal folds. The western face of Little Hangman Hill at Combe Martin exhibits such a condition very well. Throughout the whole coast which I examined, from Combe Martin to the southernmost part of Bideford Bay, the axes of the minor folds are inclined to the north in this way, so as to frequently produce the appearance of steady southern dip and greatly to increase the apparent thickness of the strata. In speculating as to the possible cause of the constant direction in which the axes of these folds are inclined, an explanation occurred to me which I believe to be the correct one, and which may be applied generally to a large number of cases. As I have been unable to find it in the works I have been able to consult, it may be desirable to publish it, since it may aid considerably the elucidation of the true structure and history of complicated districts.