A little consideration shows that the very conditions of high temperature and viscosity, which render untenable the hypothesis of a rigid obstacle, suffice to afford a full explanation of the retardation of the roots of the folds. For directed translatory movements cannot be transmitted through a fluid, pressure in which is necessarily hydrostatic, and must be exerted equally in every direction. And this applies, not only to a fluid, but to a body which will yield viscously to an impressed force. There will be a gradation, according as viscosity gives place to rigidity, between the states in which the applied force resolves itself into a purely hydrostatic pressure, and in which it is transmitted through the material as a directed thrust. The nature of the force, in the most general case, of course, has to be considered; whether it is suddenly applied and of brief duration, or steady and long-continued. The latter conditions alone apply to the present case. It follows from this that, although a tangential force 157 or pressure be engendered by a crustal movement occurring to the south, and the resultant effects be transmitted northwards, these stresses can only mechanically affect the rigid parts of the crust into which they are carried. That is to say, they may result in folding and crushing, or horizontally transporting, the upper layers of the Earth's crust; but in the deeper-lying viscous materials they must be resolved into hydrostatic pressure which may act to upheave the overlying covering, but must refuse to transmit the horizontal translatory movements affecting the rigid materials above. Between the regions in which these two opposing conditions prevail there will be no hard and fast line; but with the downward increase of fluidity there will be a gradual failure of the mechanical conditions and an increase of the hydrostatic. Thus while the uppermost layers of the crust may be transported to the full amount of the crustal displacement acting from the south (speaking still of the Alps) deeper down there will be a lesser horizontal movement, and still deeper there is no influence to urge the viscous rock-materials in a northerly direction. The consequences of these conditions must be the recumbence of the folds formed under the crust-stress, and their _deferlement_ towards the north. To see this, we must follow the several stages of development. The earliest movements, we may suppose, result in flexures of the Jura-Mountain type--that is, in a 158