were obviously at one time connected with the main mass C-the dotted line representing the conditions of surface that formerly obtained. In a word, the detached masses have been left behind during the retreat of the escarpment to its present position. The course of the River Thames, whose head-waters rise on the east side of the Cotswold Hills, was doubtless deter

mined by the inclination of the original surface of the ground. It will be observed that, like the streams represented in Fig. 16, this river flows across the outcrops of the Jurassic and Cretaceous strata, cutting through the Chalk escarpment between Wallingford and Reading.

Although wide regions may be built up of strata

FIG. 20.

SECTION ACROSS THE WEALDEN AREA. (Ramsay.)

a, Upper Cretaceous strata; b, Lower Greensand, etc.; c, Weald clay; d, Hastings sands, etc.

dipping continuously in one direction, yet it is more usual to find the direction of dip changing. Such changes may occur at wide intervals, or they may succeed each other within narrower limits. Sometimes we may have the beds of a broad area arranged in

one single anticline or syncline as the case may be. In other places the undulations of the strata may be numerous. Many examples of such structures might be cited from the rocks of Great Britain. Restricting attention for the present to gently inclined and undulating strata, we encounter a fine illustration of a broad anticline in the Chalk Downs and the Weald. (Fig. 20.) The latter might be described as a wide amphitheatre, open to the sea on the east, but surrounded in all other directions by bold bluff-like hills. Here the configuration has had precisely the same origin as the escarpments of the Midlands. The North and South Downs coincide with the outcrops of the Chalk, while the enclosed low grounds have been excavated out of underlying argillaceous and other unconsolidated strata. The Chalk, one cannot doubt, originally extended over the whole of the Wealden area, as shown by the dotted lines in Fig. 20. That high ground formerly existed within this area is clearly indicated by the fact that the escarpment of the Downs has been sawn across by streams flowing out from the heart of the Weald. Obviously when these streams first began to flow, the waterparting in the axis of the Weald must have been at a higher level than the present summit of the Downs. The whole surface has been lowered by epigene action-the less readily reduced rocks and rock-structures forming as usual the most prominent features in the landscape.

The denudation of a broad anticline composed of

harder rocks intercalated among a series of more yielding strata results, as in the Wealden area, in the formation of lines of escarpment facing each other. In the case of a denuded syncline of similar strata escarpments are likewise developed, but their faces are now turned in opposite directions. Fig. 21 shows the geological structure of a portion of Ayrshire. Here we have a series of hard volcanic rocks (v2), old lavas, in fact, intercalated between underlying and overlying sedimentary strata-chiefly sandstones and shales. The result is the same as in all the cases already considered-the more durable rocks crop out

FIG. 21.

SECTION ACROSS PERMIAN VOLCANIC BASIN, AYRSHIRE. c, Carboniferous strata; v, volcanic rocks; p, Permian sandstones.

strongly and form escarpments, but these look away from and not towards each other.

In regions which have experienced much denudation, gently inclined strata, when arranged in a series of anticlines and synclines, not infrequently give rise to an undulating surface. But this surface does not coincide with the deformations of the rocks below. In point of fact, anticlines are not infrequently represented at the surface by depressions, and synclines by elevations. These phenomena are best developed when beds or masses of durable nature are intercalated in a series of more yielding rocks. In the accom

panying section (Fig. 22) it will be observed that synclines coincide with hills, and anticlines with valleys. This configuration has been determined by the geological structure. In each hill we have practically two escarpments placed back to back. The beds h h are relatively harder than others in the series. Had no such beds occurred the synclines would probably not have been so strongly emphasised by elevations. But the presence of one or more hard beds in series of undulating and relatively soft strata does not necessarily give rise to synclinal hills. The hard beds in such a series would no doubt in time crop out at the surface

FIG. 22.

SYNCLINAL HILLS AND ANTICLINAL VALLEYS. s s, synclines; a a, anticlines; hh, relatively hard beds.

and project above the base-level of the district; but if in the synclinal troughs they descended below that level, they could have no influence upon the surface. Thus in the section (Fig. 23) a relatively hard bed crops out and forms escarpments at e e, but it descends below the base-level, b b, in the two synclinal troughs (s1 s2), which remain unaffected by it. In the third trough (s), however, it remains above the base-level, protecting the underlying softer beds, and thus forming a hill.

When a series of undulating strata contains no intercalated hard beds, but is of much the same consistency throughout, the synclines still offer the stoutest resistance to denudation, anticlines being relatively weak structures. In the former the strata are not liable to be undermined and displaced by the

FIG. 23. ESCARPMENT HILLS AND SYNCLINAL HILL.

e e, hard bed; s1 s s3, synclinal troughs; b b, base-level.

action of springs. In the latter, however, the strata hang away from the axis, and water percolating through them, and coming out along the beddingplanes, tends to their demolition. But this is a matter which will be considered more fully when we come

FIG. 24.

SECTION ACROSS WEST LOMOND HILL AND THE OCHILS. a, igneous rocks; b, red sandstones, etc.; c, basalt.

to consider the surface-forms yielded by steeply inclined and highly folded strata.

In regions long exposed to denudation all weakly built hills tend to disappear. Hence in such countries anticlinal hills are of very rare occurrence. Now and again they do occur, but only when they happen to be composed of more durable rocks than those which