the tightening-screw, the centre and plate generally rising as a dome.

The result may be described as a peripheral overfold-dome, the domical fold rolling over, with beautiful regularity, during the screwing up of the band.

The overfold-dome is well exhibited in the view (Plate XXX., fig. 1).

The perfection and harmony of the movement of the clay plates in adapting themselves to the new conditions and complicated bendings and bucklings are shown in fig. 2, which is a saw-cut section along a line a little oblique to the shorter axis of the oval.

This experiment proves very clearly that the centripetal pressure, and consequent reduction of the radius and circumference of the original discs, are met in two ways: the radii, being shortened, arch up in dome form, and the reduction of the circumference brings in peripheral folding. The result is a spirally folded dome.

Explanation of Plate XXX.-Fig. 1. View of model as taken out of the compressor, showing the peripheral overfolding of the dome.

Fig. 2. Saw-cut section across the dome looking from the same direction, or roughly parallel to the overfolding. Final size of base, 71⁄2 in. × 61⁄2 in.; height, 33 in.

CHAPTER XVI

WHAT THE EXPERIMENTS IN CIRCUMFERENTIAL COMPRESSION TELL US

THE

HE experiments described in the previous chapter enable us to understand the effect of multilateral pressure upon an assemblage of diverse and bedded strata.

It has been shown clearly by more than one geologist that such multilateral or, as I term it, circumferential compression has taken place in nature. From these diverse lateral pressures some extraordinary effects have resulted, such as the spiral twisting or screw movements of bedded rocks, similar to those recorded by Mrs. Gordon, D.Sc., in her very clever paper on The Torsion Structure of the Dolomites.'

I have, I think, shown both in 'The Origin of Mountain Ranges' and in this work that all the tangential movements of compression in the earth's crust are, in a greater or less degree, multilateral.

From whatever cause this multilateral compression arises, it is guided in its effects by the existence of lines of resistance and of weakness. We have seen that the provision of some simple contrivance to 1 Q. J. G. S. 1899, pp. 560-€33.

1

give what I have called an 'initial bias' to the movements often makes a wonderful change in the form of the model that comes from the compressor.

In the case of the Dolomites and similar mountain structures the multilateral compression is, as we may say, focussed or concentrated in a comparatively small area, though it may originate in a large one.

The experiments have satisfied me that pressure in three directions at right angles to each other for this is what the compression resolves itself into is competent, under conditions that may easily happen in nature, to produce a vast variety of unexpected forms. These forces may be concentrated in a small area, producing a rocky whirlpool, or may be spread out and taken up by long folds, which may seem to be the result of simple unilateral or bilateral pressure.

When we investigate the individual folds of strata which have been subjected to lateral pressure, we find them in their undisturbed forms to be, almost without exception, boat- or canoe-shaped, proving, as I have frequently pointed out, that the pressure producing them has acted centripetallyotherwise multilaterally.

The axis of a mountain range is seldom or never in a straight line. It curves and sweeps round to right or left, as the case may be, the effect of forces acting in the direction of the major axis simultaneously with those acting transversely; or it may be, and often is, crescentic in plan. This is

the general rule, but exceptions can be quoted where the lateral pressure has acted circumferentially more equally from the surrounding area. Stefani, in his monograph on the Apuan Alps, says that the most simple folds in their more regular courses answer to ellipsoids, and the horizontal sections of their strata to ellipses, but that the folds often bifurcate and ramify with the utmost irregularity. The horizontal course of a fold may deviate in all possible directions up to right angles. It may turn to the same part of the horizon from which it starts, and describe a semicircular, or horseshoe, or any other curve, even having a very short radius. In a vertical direction. a fold may divide in every possible way-that is, be inverted on one side of a region, or in an opposite direction in another, or elsewhere become vertical.

The predominating direction of the folds of the Apuan Alps is, however, in the direction of the meridian, where the prevailing compressions show themselves; therefore the compressing forces had the principal, but not the exclusive, direction of the parallels between east and west.1

If the lateral forces have a dominant parallel direction, the axis of the range will be approximately at right angles to them. Once a linear folding or corrugation is established, it strengthens the strata enormously as against an end-compressing force; hence the distortion of the individual folds in the efforts of end compression. A careful

1 Le Pieghe degli Alpi Apuane, Firenze.