Returning towards Sandown we see the strata dipping less and less steeply, till near the Granite Fort the rocks on the shore are horizontal. Continuing our walk past Sandown to Shanklin we pass the same succession of rocks we have been looking at, but in reverse order, and sloping the other way. It is not very easy to see this at first, for so much is covered by building; but beyond Sandown we see Sandstone Cliffs like the Red Cliff again, the strata dipping gently now to the south, and in the downs above Shanklin we see the chalk again. So we have the same strata north and south of Sandown, forming a sort of arch. But the centre of the arch is missing. It must have been cut away. We saw that the land was all being eaten away by rain and rivers. Now we see what they have done here. Go up on to the Downs, and look over the central part of the Island. We see two ranges of downs running from east to west,--the Central Downs of the Island, a long line of chalk down 24 miles from the Culver Cliff on the east to the Needles on the west; and the Southern Downs along the South Coast from Shanklin to Chale. In the Central Downs the chalk rises nearly vertically, and turns over in the beginning of an arch towards the South. Then comes a big gap, and the chalk appears again in the Southern Downs nearly horizontal, sloping gently to the south. The chalk was once joined right across the central hollow, where now we see the villages of Newchurch, Godshill, and Arreton. All that enormous mass of rock that once filled the space between the downs has been cut away by running water. An arch of strata like this [Inverted-U], such as the one we are looking at, is called an _anticline_. When the arch is reversed, like this [U], it is called a _syncline_. Looking north from the Central Downs over the Solent we are looking at a syncline. The chalk, which dips down at the Culvers and along the line of the Central Downs, runs like a trough under the Solent, and rises again, as we see it on the other side, in the Portsdown Hills. We might suppose the top of an anticlinal arch would be the highest part of the country; that, even if rain and running water have worn the country down, that would still stand highest, and be worn down least. But there are reasons why this need not be so. For one thing, when the horizontal strata are curved over into an arch, they naturally crack just at the top of the curve, so and into the cracks the rain gets, and so a stream is started there, which cuts down and widens its channel, and so eats the land away. Again, the rising land only emerges gradually from the sea, and the sea may cut off the top of the arch before it has risen out of its reach. Moreover on the higher land the fall of rain and snow is greater, and the frosts are more severe; so that it is just there that the forces wearing down the land are most effective. [Illustration: curve with two v-shaped marks at center]