contains subangular blocks of Millstone Grit about 3 feet in diameter. This gravel continues some way up stream.

Opposite Lud Stream islands there is a gravel in a similar position and apparently of a similar character, but the section is not clean cut. It is surmounted by another bank of gravel (section also not good) apparently of the same general character, but containing much larger subangular blocks of grit. The top of this gravel bank is 75 feet above the river, and from its edge there stretches a gently rising and undulating surface for about one sixth of a mile to the foot of the hill called South Nab. This plateau may be entirely of gravel. On the left bank of Posforth Beck, a little below the first bend in the stream above its junction with the Wharfe, there is a section in true glacial drift. The material consists of angular debris of grits and limestones, and the blocks of the latter are well glaciated. It contains also fine stratified sand, beautifully false-bedded. The boulder-bearing drift seems to wrap round the sand, but the section is not clear. Further down stream there is a similar angular deposit. There are plenty of rounded stones; but the characteristic feature is angularity.

Pickles Gill, above its junction with Tom Taylor Dike, is full of drift, consisting of angular blocks and pebbles of grit and limestone. The pebbles are mostly of limestone; and the limestones are glaciated and scratched, particularly the pebbles

It is not necessary to give any more details of sections, as it would merely be a repetition of what has been already said. The general result is that the boulder beds vary from a true boulder clay to a stony mass, which consists sometimes of a heap of angular fragments, while at others it contains a large number of pebbles; and that in the latter case, on the Millstone Grit area in the south, the larger proportion of the pebbles are of limestone, and that the limestones are generally scratched, the grits but rarely so. Evidently the reason why most of the pebbles are of limestone is that the limestones have come the greater distance.

Further, the true glacial boulder beds are mixed up with finely stratified sand and gravel. Lastly, well water-worn gravels, containing large boulders and angular blocks, line the valley sides.

C

in terraces, like ordinary river terraces. This makes me think that there is no real distinction to be drawn between the older river gravels and the water-worn gravels with boul lers, which are generally set down to the glacial period as something quite sui generis. I believe on the contrary that the ordinary river gravels run back to glacial times, and gradually merge into the glacial deposits.

I will now consider the solitary case of foreign blocks occurring in the valley of the Wharfe. These consist of Silurian Grits like some of the Silurian Rocks that occur in places in Ribblesdale. In Wharfedale these Silurian boulders are confined to the portion of the valley between Chapel House Lodge and Burnsall, that is, all that I have met with are below Chapel House Lodge and above Burnsall. They are most plentiful about Linton and Threshfield. The greater number of them have long since been cleared off the ground and built into the walls for "throughs," where their remains may still be seen; but some few I found still lying about on the surface of the land, and some may be seen in the drift cut through by becks descending from Threshfield Moor. I could find noue above the general level of the drift of the more open country, which near Threshfield reaches about the height of 800 feet above sea level. I examined the country lying between the site of the boulders and the outcrop of Silurian Rocks in Ribblesdale; but though there are plenty of Millstone Grit boulders lying on the bare surface of the limestone, not a single Silurian boulder was to be found. It was quite clear that these Wharfedale boulders had not come over the fells from Ribblesdale. Whence and how then did they come? It may be thought at first sight that they came on floating ice discharged from a glacier debouching at the mouth of Ribblesdale near Settle. But there are great difficulties connected with this view, the chief of which is this: if these boulders drifted eastward from Settle on floating ice, when the sea level was but little higher than the present 800 contour line, and Wharfedale was a fjord, whether filled with ice or water, how did they surmounting the rock barrier at Nether side manage to travel up the dale or fjord towards Chapel House? for assuredly ice, or water carrying ice, was flowing down Wharfedale all the while. This is an insuperable difficulty to the floating ice theory.

I believe that the true explanation is this: At the foot of Kilnsey Crag strong springs break over from the limestone rock, just above the level of the alluvium; and also in Littondale strong springs break out at the foot of the limestone scars, the hillside below being formed of a mass of detritus, which as completely conceals the underlying rocks as the alluvium at Kilnsey does. Now, seeing that strong springs break out at the foot of scars formed by limestone of great thickness, and that it is only a little way below the springs that we find Silurian boulders in the dale, it seems likely that the springs are thrown out by Silurian rocks in places in the bottom of the valley, hidden from sight by superficial detritus, drift and alluvium, and that our boulders were thence derived. This makes everything simple; and the thickness it allows to the limestone is quite equal to the thickness of the limestone in Ribblesdale..

In conclusion let me say that thus far the drift phenomena of Wharfedale lend no support to the theory that the whole country was once overridden by an ice cap descending from the pole. There is no evidence here of foreign ice; but everything is in favour of huge confluent glaciers, or ice-sheets if you like, of home-made ice. Nidderdale, too, supports the same conclusion. There are no foreigners there at all.

THE GLACIATION OF YORKSHIRE.

BY PERCY F. KENDALL, F.G.S.,

LECTURER ON GEOLOGY AT THE YORKSHIRE COLLEGE.

The address, of which the following is a rather free rendering, was delivered at very short notice. In its present form it is designed rather with a view to directing the attention of geologists to some of the larger problems connected with the glacial phenomena of Yorkshire than as an attempt to offer a full or final solution of them.

I have elsewhere stated, somewhat at length, the grounds upon which I have been led to reject the theory of a great submergence of England and Wales during the glacial period, and to ascribe the phenomena to the operation of great glaciers with their concomitants of sub-glacial rivers, extra-morainic lakes, &c., acting upon a landsurface standing at about, and certainly not below, its present level.

I have also specified the course, general effects and approximate limits of the glaciers so far as I had been able, either by a study of the literature of the subject, or in the field, to determine them. I may, however, briefly recapitulate the points germane to the present enquiry.

The gradual approach of the ice-age produced at first small valleyglaciers in our great hill clusters, e.g. the Highlands and Southern uplands of Scotland, the Lake District, North and South Wales, the mountainous parts of Ireland, and perhaps some few valleys of the Pennine Chain.

With the accentuation of the cold these glaciers grew, coalesced, and extended out upon the low grounds or into the sea according to situation. Such as reached the sea detached icebergs from their tronts, and these drifting hither and thither in the random fashion so characteristic of ice floating in a tideway, would scatter their loads of boulders in a very erratic and irregular fashion over the sea floor. A further stage of refrigeration would bring about further confluences of the ice streams, and such as debouched into such * Man and the Glacial Period (Internat. Scient. Series) pp. 137-181, and Geol. Mag. Nov., 1892, p. 491.

*

shallow waters as the Irish Sea and the Frith of Clyde would be unable now to find free flotation for the bergs. The condition of the Irish Sea powerfully affected that of Yorkshire, and indeed was one of the greatest factors in determining the nature and mode of glaciation of this county, and we may now consider what would be the state of affairs there as the Glacial Period approached its climax. We know from the unimpeachable testimony of striæ upon rock surfaces that there was a general convergence of glaciers of immense size upon the northern portion of the Irish Sea from almost every point of the compass. The glaciers of Kirkcudbrightshire and adjacent Scottish counties poured into it from the northward, those of the Lake District, radiating from a centre about Great Eud,† came in from the north-east, and others from the Fells of Yorkshire and Lancashire from the east. North Wales sent its quota from the south, and, finally, the great central basin of Ireland sent a portion of its surplus in from the westward. Now, in view of the fact that the depth of the Irish Sea rarely exceeds 50 fathoms and that glacial striæ are found on hills quite close to the coast at altitudes exceeding 1,000 feet above sea level, it will to many people not be surprising to find that this great influx of ice completely displaced the sea and converted the area into a huge ice-basin, in many particulars resembling that of Greenland. Of the depth of the ice we have a criterion, though an imperfect one, furnished by the fact that the highest point of the Isle of Man (Snaefell, 2,034 feet) was completely buried beneath the ice, and, along with its companion hills, was glaciated from top to bottom. Under these conditions new directions of flow would be taken, some of which might have been predicted a priori had we not previously determined them a posteriori. Three obvious outlets from the Irish Sea suggest themselves: one through the St. George's Channel between the Wicklow Mountains and the Snowdonian massif; one by way of the broad low valley between the Pennine Chain and the Welsh hills; and a third through the North Channel between the Antrim Coast and the Rinns of Galloway. The first and second of these routes we find by the evidence of striæ, boulder

* Geol. Surv. Mem.

+ Ward, Q. J. G. S., vol. xxxi.

‡ Tiddeman, Q. J. G. S., vol. xxviii. ; Goodchild, op. cit., vol. xxx.