338 Proceedings of the Royal Physical Society. 



of the ocean. There is one circumstance in reference to the 

 formation of these terraces on the borders of lakes which 

 does not seem to be generally taken into consideration ; that 

 is, the effect produced by floating ice. Water washing 

 against a friable bank will no doubt wear it down, and in 

 course of time will form a terrace ; but the operation will 

 be much more speedily effected if it bear on its surface 

 masses of ice which are carried by the waves against the 

 banks, and act like the battering-rams of the ancient war- 

 riors. The accounts that we read of the devastations some- 

 times caused by floating ice brought down by the rivers in 

 America amply confirm these remarks. It seems to us of 

 great importance, in forming any hypothesis in regard to 

 level terraces, raised beaches, sea-margins, or whatever other 

 name may be given to them, to remember that they are of 

 two kinds. One is formed below the surface of the water, 

 and the other above it ; one is produced by the retreating, 

 and the other by the advancing billow. The deposits left by 

 rivers are altogether different in their origin and character 

 from those which we find on the banks of lakes and on the 

 shores of the sea. So long as the river is confined in a nar- 

 row channel, the gravel and mud which it brings down are 

 swept along together by the current. When the channel 

 becomes wider, the sand and gravel are left behind, and 

 form a bed, which has always a distinct downward slope 

 according to the course of the stream. When the channel 

 is still farther enlarged, the current becomes slower, the mud 

 is deposited, and forms deltas and valleys, like those of the 

 Mississippi and Ganges, with a surface nearly horizontal. 



Character and Appearance of these Terraces. 

 The terraces found on the sides of lakes are regular and 

 horizontal, making allowance, of course, for those places 

 where the waves are particularly high ; but this is not the 

 case in regard to the deposits left by the ocean. The upper 

 terraces, formed by the advancing billow above ordinary high- 

 water mark, sometimes exhibit a very remarkable horizon - 

 tality of outline. When the material thrown up consists of 

 shingle or gravel of such weight as to resist the action of 



