towards Kirk Michael, until they disappear altogether from the cliffs and the. shore. They stand up conspicuously along the ancient shore line extending from Kirk Michael to Ballaugh, Sulby, and Ramsey, commanding the low, sandy, and marshy region which forms the northern portion of the island, contrasting in its flatness with the lofty rolling Ordovician hills behind, culminating in Sartfell, Snaefell, and North Barule. This contrast is obviously the result of a difference in the physical character of the rocks in the two districts. The problem as to which rocks underlie the glacial strata in the former, which had occupied the author's mind for many years, is now partially solved by the three borings which have been made under his advice by Messrs. Craine in 1891-4 in search of the Coal-Measures of the Whitehaven field, at the Point of Ayre, at Blue Point, and at Lhen Moar. The boring at Lhen Moar revealed the existence of the Carboniferous Limestone at a depth of 167 feet 6 inches below the drift. The next bore-hole, at Blue Point, about 4,050 feet to the north-east of that at Lhen Moar, revealed the presence of more than 60 feet of Red Sandstone buried 171 feet beneath the drift. The Red Sandstone in this section is, in his opinion, identical with the St. Bees sandstone, or lowest member of the Triassic formation in the district of the Lakes.

This conclusion is greatly strengthened by the discovery in the third boring at the Point of Ayre, to the east of the lighthouse, of the Triassic marls with salt, at a distance of a little under five miles from Blue Point. The diamond drill was used from a depth of 452 feet to the bottom. The total thickness of the salt-beds amounts to 33 feet 6 inches, and the bore-hole happened also to intersect a brine run 2 feet 6 inches in depth. If this section be compared with that published by Mr. Dickenson of the saliferous marls of Duncrue, near Carrickfergus, it will be found to be practically identical. The same series of salt-bearing marls is also worked at Barrow-in-Furness and at Preesal, near Fleetwood. The salt-beds in each of these cases are variable in thickness, and those in the Isle of Man are thinner than in the other localities. It must, however, be remembered that the Manx boring has not been put down to a sufficient depth to test the true thickness of the salt-field. The discovery is of great theoretical importance, because it links on the deposit at Barrow to that of Carrickfergus, and shows that the Irish Sea was an area in which the salt-bearing Triassic marls were deposited. It points towards the truth of Mr. Dickenson's suggestion that the Cheshire salt-field was formerly continuous with that of Ireland. These marls have since been broken up, faulted, and denuded away in many places. It is an open question how far those of the Isle of Man are now continuous under the sea eastwards to Barrow and Fleetwood, and to the north-west in the direction of Carrickfergus.

All these rocks are buried under a great thickness of boulder sand, gravel, and clay, amounting at the Point of Ayre to 298 feet. To this also must be added the height of the drift hills close by, formed of the same materials, which would give the total thickness as not less than 450 feet in the extreme north. The rocky floor on which it rests dips rapidly to the north-east towards the deeper part of the Irish Sea.

The discovery of this salt-field is likely to add a new industry to the resources of the Isle of Man.

3. Strictures on the Current Method of Geological Classification and Nomenclature, with Proposals for its Revision. By Sir HENRY HOWORTH, F.R.S.

4. On the Pleistocene Gravel at Wolvercote, near Oxford.
By A. MONTGOMERIE BELL, M.A.

The section is a typical illustration of a somewhat advanced period of Quaternary time, and in its general features resembles the sections at Hoxne and Bedford, originally published by Sir John Evans, while it is very different from the implementiferous beds of an earlier age which are found on the Greensand escarpment of Kent and Surrey.

(1) The Oxford clay, the bed-rock of the district, is furrowed at the surface into wavy hollows, irregularly filled with earth containing quartz, lydian-stone, and quartzite pebbles, and bunches of unstratified gravel from the Thames basin. it is certainly a northern drift, in the writer's opinion a glacial drift.

(2) At Wolvercote this drift is invaded, and for a certain distance removed by river action, which has hollowed out the clay to a depth of 17 feet, and subsequently filled up the hollow with horizontal layers of gravel, mud, and sand. (3) The junction where the river eats into the drift is clearly visible. The bank of drift, underworn by the water, overhangs the horizontal layer.

(4) The lowest bed of the riverine deposits differs from the others: it consists of 2 feet of gravel and sand, in lenticular, shorn, and current-bedded layers, showing by the size of the pebbles a somewhat rapid current. At the very base, half embedded in the clay, many mammalian bones have been found, and six Palæolithic implements. The implements represent well-known types of the rivervalley period, and the mammoth is conspicuous among the animal remains, though Equus, Cervus elaphus, and Bison priscus are also present.

Layers of sand are intercalated with the gravel, from which eleven species of shells have been identified. They are all recent and, generally speaking, stunted in size.

(5) Above the gravel are two inches of sandy peat, marking a land surface of some duration. Nine plants were identified by Mr. Clement Reid, which are all species still to be found in the immediate neighbourhood.

(6) For 14 feet sand and mud foliow in successive layers. In these no fossil has been found. They indicate quiet river action and an increase of pluvial conditions.

(7) Towards the surface they are traversed by an irregular line of trail, which marks apparently the movement of a sludgy mass along the surface, punching it downwards by its weight.

(8) The surface level at Wolvercote is 240 feet above Ordnance datum. The adjacent river surface is 195 feet. The gravel at Somerton and Oxford is about 218 feet. Thus between the two gravels there is a distance of 22 feet. At the present rate of erosion many thousands of years would be necessary to remove 22 feet from the general surface; but the fact that the remains of man and of animals are the same in both gravels proves that they belong to a similar age, though the gravel at Wolvercote is somewhat older than the other. We must therefore consider that the denuding agents-rain and frost-were more active at that period than they are at the present day.

5. On Prehistoric Man in the Old Alluvium of the Sabarmati River in Gujarat, Western India. By R. BRUCE FOOTE, F.G.S.

Two finds of chipped (Palæolithic) implements were made in a bed of shingle occupying a definite horizon in the lower part of the old alluvium of the Sabarmati in latitudes 23° 25′ and 23° 40′ N. (about 330 miles north of Bombay). The implements were found at a depth of about 70 feet below the surface of the alluvium, which is here over 100 feet thick. The alluvium is overlaid by löss and wind-blown loam, which varies from 80 to 150 feet in thickness. The river has cut itself a bed varying from 100 to 200 feet in depth in these deposits, showing that a great interval of time must have elapsed since the deposition of the old alluvium in which the implements are embedded. On the surface of the löss, and in many cases on the summits of the blown-loam hills, Neolithic remains in the form of flint flakes and cores of the Jabalpur type were found together with fragments of archaic pottery.

6. On the Shape of the Banks of Small Channels in Tidal Estuaries. By Professor H. HENNESSY, F.R.S.

Many years since my attention was attracted by the peculiar shape of the soft mud banks bordering the channels through which water drains off the beds of,

tidal estuaries. These banks, instead of presenting sloping planes or concave surfaces, are always convex when the matter of which they are formed is soft and yielding. A section of such a pair of banks would present an approximation to a cusp in the middle portion.

This peculiar outline is manifestly due to the action of water on the yielding matter. A few years since I found a result which may assist in exploring this phenomenon. The velocity of water in a channel is well known to depend on the ratio of the area of the cross-section to its perimeter. This ratio is variable in all known channels of water flowing through rigid materials such as canals and rivers. On investigating the form of section corresponding to a constant ratio of the quantities referred to, I found that it would be represented by a pair of catenaries with their ends meeting so as to form a cusp like that in the estuary channels.1 The result would be that, with every depth of water in such a channel, the flow would have nearly the same velocity.

7. Report of the Committee on Earth Tremors.-See Reports, p. 145.

8. Interim Report of the Committee on the Investigation of a Coral Reef.

9. Report of the Committee on Underground Waters.-See Reports, p. 283.

10. Report of the Committee on the Marine Zoology of the Irish Sea. See Reports, p. 318.

11. On a Keuper Sandstone cemented by Barium Sulphate from the Peakstones Rock, Alton, Staffordshire. By W. W. WATTS, M.A., F.G.S.

Professor F. Clowes has described a sandstone from the Himlack Stone, near Nottingham, in which the grains are cemented with crystalline barytes, the amount of this material varying from 28 to 50 per cent. in different specimens. This rock occurs at the base of the Keuper Sandstone of that locality. A somewhat similar rock, occurring at about the same horizon, is described by Mr. A. Strahan,3 from Beeston Castle in Cheshire, and the same author refers to the frequent occurrence of barytes in the Keuper breccias.

Bearing these facts in mind, the writer visited a curious isolated stack of rock, called the Peakstones Rock,' near the village of Alton in Staffordshire, which is figured in Professor Hull's memoir on The Triassic and Permian Rocks of the Midland Counties of England.' This stack is made of the lower beds of Keuper Sandstone, but its outer portion has lost whatever cement it may once have contained. It is, however, situated at the end of a spur which projects into a valley, and exposes a good deal of bare rock. This rock contains what at first look like several veins of barytes two or three inches thick, striking along the spur and straight through the place occupied by the Peakstones Rock. On examination of

Proceedings of the Royal Society, vol. xliv. p. 108.

2 Rep. Brit. Assoc., 1885, p. 1038; 1889, p. 594; 1893, p. 732; and Proc. Roy. Soc., vol. xlvi. pp. 363-369.

1 Mem. Geol. Survey. Exp. Quarter Sheet, 80, S.W., p. 7.

specimens the veins are seen to be planes along which the sandstone is cemented by barytes. The specific gravity of the rock is 3'09, and, as the grains are chiefly subangular fragments of quartz and felspar, it must contain about 28 per cent. of barytes. This almost insoluble cement has undoubtedly given rise to the spur above alluded to, and almost as certainly has caused the survival of the Peakstones Rock, which now, however, is so much exposed to the weather on all sides, and both to mechanical and chemical disintegration, that if any cement is still left it can only be in the inner part of the mass which cannot be reached by ordinary means. Another specimen from West of Kent Green, near Congleton,' containing barytes, and with a structure very like that described by Mr. Strahan, was also referred to. The paper was illustrated by a set of photographs which the author owed to the kindness of Mr. A. A. Armstrong and Mr. P. Simpson.

12. Report of the Committee on the Volcanic Phenomena of Vesuvius. See Reports, p. 315.

SECTION D.-BIOLOGY.

PRESIDENT OF THE SECTION.-Professor I. BAYLEY BALfour,
M.A., F.R.S., F.R.S.E.

THURSDAY, AUGUST 9.

[For the President's Address see below.]

The following Reports were read :

1. Report on Investigations made at the Zoological Station, Naples. See Reports, p. 335.

2. Report on Investigations made at the Laboratory of the Marine
Biological Association, Plymouth.-See Reports, p. 345.

3. Report on the Zoology of the Sandwich Islands.-See Reports, p. 343.

4. Report on the Fauna and Flora of the West India Islands.
See Reports, p. 344.

5. Report on the Index Generum et Specierum.-See Reports, p. 347.

The President delivered the following Address:

THE prospect of visiting Oxford to-day has, I am sure, been to all of us a pleasant one, and we who are specially interested in biology have looked forward to our meeting at this time with the distinguished members of the Oxford Biological School. But as we gather here there will, I think, be present to the minds of all of us a thought of one member of that school, whom we had hoped to meet, who is recently gone from it in the prime of his intellectual life. By the death of George John Romanes biological science is bereft of one of its foremost expositors, Oxford is deprived too soon of one whose mental power was yet in its zenith, and each one of us who knew him cannot but feel a deep sense of personal loss; and we shall in our meeting here sadly miss the man brimming with a geniality which robbed differences of their difficulty and charmed away bitterness from those controversies in which he revelled. This is not the occasion upon which to dwell on his character, his merits, or his work. We must all, I think, have appreciated the graceful accuracy with which these were sketched in the pages of 'Nature' by one of his colleagues; but under the shadow, as we are here, of his recent death, I