THE GEOLOGY OF BERMUDA. 25 



deep and wide gulf. These hollows are simply caves whose roofs liave 

 been eroded by the mechanical and chemical action of water, until, too 

 -weak to support themselves, they have caved in. On the walls of 

 these unroofed caves beautiful stalactites may be seen half enveloped 

 with velvety mosses and feathery ferns-^a strangely beautiful combi- 

 nation of the adornments of the underworld with those of the world 

 of daylight. In other cases the process of erosion has continued still 

 further, so that the sides of the cave, as well as the roof, have been 

 entirely removed, and nothing is left to mark the site of the former 

 cave but a floor of crystalline stalagmite. Near Walsingham on the 

 Main Island, and near Mullet Bay on St. Georges, I observed consid- 

 erable areas where tbe coarsely crystalline calcite forming the sur- 

 face rock is unquestionably a stalagmite floor — the only memorial of a 

 former cave. 



THE "RED EARTH." 



The so-called " red earth " bears striking testimony to the amount of 

 erosion which the islands have undergone. The usual superficial soil 

 of the islands is a clayey earth, sometimes of a deep brick-red color, 

 sometimes showing various shades intermediate between this deep red 

 and the white or cream-color of the underlying rock. The material is 

 occasionally somewhat firmly consolidated, but usually quite soft and 

 earthy. It varies much in depth, forming deep pockets in some places, 

 while in other places the white rocks are bare. It often occurs in 

 cracks and cavities in the rocks. Where any considerable thickness of 

 the drift-rock is exposed in a section, as at the extensive quarries on 

 Ireland Island, one or more layers of the same "red earth" may gener- 

 ally be observed extending nearly horizontally at intervals through the 

 rock. Various unsatisfactory explanations of the origin and nature of 

 this "red earth" have been given. Jones formerly believed it to be 

 "composed of decayed vegetable matter";* and this is indeed the 

 common opinion of the inhabitants of Bermuda. Nelson conjectured 

 that it was largely derived from the excrements of bats and birds, t 

 The true explanation of its origin is undoubtedly that given by Thomson, 

 as follows: "The coral-sand, like the mass of skeletons of surface ani- 

 mals accumulated at the bottom of the ocean, does not consist of car- 

 bonate of lime alone. It contains about 1 per cent, of other inorganic 



" On the Geological Features of the Bermudas : in Proceedings and Transactions of 

 the Nova Scotian Institute of Natural Science, Vol. I., Part IV., Art. II., 1867, p. 21. 



t Dana, Corals and Coral Islands, p. 391. The citation is from a paper on the Baha- 

 mas, ill Quarterly Journal of the Geological Society of London, 1853. 



