io8 GEOLOGY OF THE ISLE OF WIGHT 



Chine is a long winding ravine between steep walls, the 

 stream at the bottom making its way through blocks of 

 fallen strata. 



The cause of these chines seems to be the same in all 

 cases. It may be noticed that Shanklin and Luccombe 

 chines are cut in the floors of open combes, wide valleys 

 with gently sloping floors ; and at each side of these chines 

 is to be seen the gravel spread over the floor of the old 

 valley. It can scarcely be doubted that these combes are 

 the heads of the valleys of the old streams, which flowed 

 down a gradual slope till they joined the old branch (or, 

 rather the old main river)* of the Yar, flowing over land 

 extending far over what is now Sandown Bay. When the 

 sea encroached, and cut into the course of this old river, 

 and on till it made a section of what had been the left 

 slope of the valley, the old tributaries of the Yar now fell 

 over a line of cliff into the sea. They thus gained new 

 erosive power, and cut back at a much greater rate new 

 and deeper channels ;^with the result that narrow trenches 

 were cut in the floors of the old gently sloping valleys. 

 The chines on the S.W. coast are to be explained in a 

 similar way. They have been cut back with vertical sides, 

 because the encroachment of the sea caused the streams 

 to flow over cliffs, and so gave then power to cut back 

 ravines at so fast a rate that the weathering down of the 

 sides could not keep pace with it. The remarkable wind- 

 erosion of these bare south-westerly cliffs by a sort of 

 sand-blast driven before the gales to which that stretch 

 of coast is exposed has already been referred to. 



A few words in conclusion to the reader. I have tried 

 to show you something of the interest and wonder of the 

 story written in the rocks. We have seen something of the 

 world's making, and of the many and varied forms of life 

 which have succeeded each other on its surface. We have 

 had a glimpse of great and deep problems suggested, which 

 * Seep. 91. 



