shaves away the edges and grinds down the faces of stratified rock-masses. It can produce an undulating surface; but the vertical extent of the heights and hollows must be small in comparison with their breadth. It cannot scoop out small deep basins, deep grooves pointing in various directions, and, least of all, funnel-shaped holes running down through slabs of rock. Water finding its way through cracks in glaciers may perhaps produce all these forms of rock-surfacein valleys; but on the summits of limestone hills such phenomenamust be referred to the action of rain or sea-waves. Rain (as I have already shown in this Journal[1]) cannot grind rock-surfaces so as to leave smooth curvilinear hollows cutting continuously through hard and soft parts of the stone; but it must communicate a rough irregular surface minutely corresponding to variations in lithological structure or composition.
b. Rock-work of Birkrigg Moor and Hampsfell.—On the rocky summit called Birkrigg Moor near Ulverstone, and on the still higher eminence called Hampsfell, near Grange, rain is now clearly breaking up everything like regularity of form. It is converting previously smooth basins into rough, pitted, and fretted depressions, and smooth grooves into jagged stone gutters. But these hills, especially the borders of Birkrigg Moor, present proofs of the most undeniable kind that no agent has formed the smooth and regular rockwork since the Glacial period. The basined, grooved, and funnelled surfaces of limestone rock run under drift without any change of form, excepting that, under damp clay, in many places the surface has been roughened. The connexion between sculptured limestone rock-surfaces and drifts in this district offers a new field of research to the geologist. So far as I have observed, the hollows must nearly all have been ground out before the deposition of the Upper Boulder-clay. They not only run under a loose reddish drift on the borders of Birkrigg Moor, but under the decided upper drift further westward;and sculptured boulders may be seen buried in this drift. Large sculptured boulders on Stainton Green (see sequel) present grooves and other hollows running continuously from the upper to the under surface, and along the latter down into the matrix of red loam[2]. The semi-island of Dunnerholme (see sequel) is covered with an upper drift which rests on sculptured limestone—the rounded stones of this drift still sticking in the funnel-shaped holes they apparently helped to grind out. But the period of the sculpturing of the limestone rocks must be carried still further back; for sculptured boulders may be found in hard Lower Boulder-clay, though in this clay the sculptured surface, the typical form of which can be at once recognized, has become irregularly pitted. Under the Lower Boulder-clay the rock-surfaces often present the same general forms, though in most places they have become fretted through some kind of chemical action, probably intensified by the humus carried down from the surface by rain-water percolating through crevices.
