In the following pages the writer has endeavoured to consider this skeleton in a common-sense way, and to arrange the limbs with reference to ordinary mechanical principles, and also by comparison of the bones with those of different mammalian and reptilian types. A good many years ago, Von Meyer was so struck by the colossal and rather straight hind-limb bones of the Dinosauria and their superficial resemblance to those of elephants, that he proposed the term Elephantopoda; but at that time the group had not been classified by Marsh and others into distinct sub-orders, with very different limbs.

Diagnosis.—Asty incrusting, uniserial, with bilateral and unilateral branching; œcia dimorphic; normal œcia with short caudæ or without caudæ pyriform; termen a complete, high, narrow, oval ridge with few small spines on its circumference and, at the proximal-lateral corners of the aperture, a pair of stout spines which sometimes, if not always, bend towards one another and fuse in the middle line; apparently there are no spines around the distal end of the aperture; extra-terminal front-wall well-developed proximally, and arched; intra-terminal front-wall a wide, depressed lamina; aperture oval, sub-quadrate, somewhat constricted laterally; avicularia small, one, or a pair, placed laterally and somewhat distally with regard to each aperture, rather abruptly pointed with the distal ends curved towards the aperture they encompass; ovicells hyperstomial.

There are many features characterizing the surface portions of deposits which are very puzzling. Indeed, the more one studies them the less one is inclined to be dogmatic. Many true glacial deposits show remarkable signs of disturbance throughout their mass. We may instance the Contorted Clay of Norfolk as a case in point. Even when boulder-clays are too uniform in texture to show signs of movement they are often kneaded into the rocks upon which they rest, and rocks of all ages often show such disturbances when they have been overridden by ice. Here we are dealing with the results produced by the action of forces operating from without the deposit.

The existence of two ridges of Pre-Cambrian rocks in South-West Carnarvonshire is well known. The more northerly of these ridges extends from near Bangor to Carnarvon, a distance of just over 8 miles. It runs in a south-westerly direction, close to and parallel with the Menai Straits, and terminates at the hill called Twthill—a prominent feature in the town of Carnarvon. The more southerly ridge—commonly known as the St. Annes–Llanllyfni ridge—stretches from St. Annes in the Nant-Ffrancon Valley to the village of Llanllyfni in the Nantlle Valley. This ridge is 12 miles long and has a south-westerly trend; it is therefore roughly parallel to the Carnarvon-Bangor ridge, from which it is distant about 2 miles. It is with the area at the south-west end of the St. Annes–Llanllyfni ridge that the following remarks are concerned. The object of my

investigation has been to establish the succession of rocks in this area, and to correlate it, if possible, with the main portion of the ridge lying to the north-east. The results are set out in detail in the following pages. Text-map I shows the position of these two Pre-Cambrian ridges and also the extent of the area to be considered in detail.

By far the greater part of the volcanic rocks of Malaya consists of fragmental deposits, which at first sight seem to be andesite-tuffs and andesite-breccias, for most of them contain abundant fragments of andesitic lava. However, the majority also contain numerous fragments of quartz, some of which occur as isolated angular grains in the cement, others embedded in a very fine-grained siliceous rock, and sometimes showing rounded outlines and even bays, invaded by the siliceous aggregate. Tuffs which occur in certain localities, e.g. at Kuala Tekal, Tembeling, and Sibah near Kuala Lipis on the Pahang railway, contain the usual fragments of andesite-lava, with numerous fragments of quartz like those just described, and in addition fragments of rhyolite-lava or quartz-porphyry. It is probable that the great majority of the Pahang Volcanic Series of tuffs are formed of an admixture of andesitic, rhyolitic, and sedimentary material.

At a time when the resources of every branch of science are being devoted to the furtherance of the War, it is not inappropriate to consider in what way geological research is being turned to account. At first sight it might appear that the work of the stratigraphical geologist, the palaeontologist, or the petrographer might be of domestic and academic interest, but would be unlikely to influence the course of a worldwide war. Such researches have received a respectful toleration in this country, but have been regarded as a luxury, to be abandoned first among luxuries in time of stress. I venture to hope, however, that it will not be so necessary in the future as it has been in the past to expend energy in urging the claims of science. In the foundation of the Imperial Trust for Scientific Research we may recognize that at least one good result has arisen from the calamity of war.

In the course of an investigation into British resources of sands and rocks for glass-making and refractory purposes, which is being carried out by my colleague Dr. P. G. H. Boswell, it became desirable to search for an alumina-bearing siliceous rock low in iron-content. Among other rocks considered by Dr. Boswell was the granophyre of Brandy Gill, and a sample, obtained from Mr. W. Hemingway, was submitted to Dr. H. F. Harwood for analysis. So far as industry is concerned, the analysis indicates that the iron-percentage of the rock is too high for its use in glass manufacture. As such an analysis would be of interest and value to geologists—there being few first-class analyses of Lake District igneous rocks—and as it seemed undesirable that only a bare record should occur in a technological publication, Dr. Boswell handed to me the analysis and a sample of the granophyre, with the suggestion that the rock might be described in the pages of the Geological Magazine.

During the past seven years I have published several papers based upon preliminary investigations on the anatomy of the Holectypoida, an order of the Euechinoidea which, by general consent, is regarded as transitional between the Regular and Irregular sections of the class. Continued study, aided by increased mental and technical experience, has revealed an ever-increasing number of interesting features in the order, and has made possible more definite and confident surmises as to its affinities with other groups. The series of contributions here commenced is intended to amplify, and in some cases to correct, the statements made in my earlier papers. With very few exceptions the actual descriptions that are given in the former work still seem to be essentially correct, but every fresh examination, even of the same specimens, testifies to the incompleteness of most of the observations hitherto published. New material, better or differently preserved, often draws attention to features overlooked or discarded as unimportant in the examples already studied. No apology is needed for the publication of incomplete descriptions of natural objects, since complete knowledge is as yet nowhere attained. When inaccuracies of observation have been detected, their nature and cause will be stated frankly, accompanied, verbally or not, by suitable contrition.

When fish-remains were first discovered in the Ludlow bone-bed and other horizons of the Upper Silurian series, some of the fragments were regarded as toothed jaws by Agassiz, who described them under the names of Plectrodus mirabilis, P. pleiopristis, and Sclerodus pustuliferus. The same fossils were afterwards considered to be of Crustacean origin by M'Coy, but definitely proved to be fish-remains by the microscopical examination of Harley, who pointed out that while they could not be teeth or jaws, they appeared to him to be “the posterior spines of the cephalic plate of some Cephalaspidian fish”. The latter view was adopted by Lankester, who treated the specimens as parts of the cornua of a small Cephalaspidian headshield which he named Eukeraspis pustuliferus.

To the student of Cambrian geology, the writings of C. D. Walcott are always matters of enlightening study, not only for the matter they contain but also because they combine that admixture of stratigraphy and palæontology in which the latter, though accorded a prominent position, is always used to the full to subserve the wider claims of the former.

In the opening article of the Geological Magazine (Vol. I, p. 5, 1864) Mr. J. W. Salter remarks: “An obscure but novel group of organic remains comes to light in some well-worked district for which we have as yet no fixed geological place,” and this description may well apply to the fauna dealt with in the following pages.