THE FLORIDA REEFS. 87 



upon, and their absence is readily accounted for by the con- 

 stant disturbance of the shore-line deposits, which reduce httle 

 by little the larger fragments of shells and corals, or echino- 

 derms, to a breccia, or again to oolite or fine sand. This is 

 nowhere so well seen as on the shore line of Key West to the 

 north of Fort Taylor. (Fig. 53.) There the outer reef is suffi- 

 ciently distant to allow waves of considerable size to break upon 

 this coast, and then strike to a low line of shore rocks. These 

 rocks are completely riddled by larger or smaller cavities made 

 by boring mollusks, annelids, sea-urchins, etc., or left by fossils 

 or fraofments of corals which have fallen out. Thus weakened, 

 large masses are easily undermined by the water, which washes 

 around them with considerable force. They fall off, become 

 then broken into smaller and finer pieces, which are again re- 

 ground in their turn, and are finally either re-soldered into finer 

 breccia or coarse oolite, or into the finest oolite or sand, accord- 

 ing to the composition of the rock. This is then cemented 

 again to the shore line, forming a new line, more or less 

 regularly stratified, dipping towards the sea, and, when ex- 

 posed to the action of the air, soon coated with a thin film of 

 hard limestone. This hardens, and forms the ringing crust of 

 the rocks found everywhere on the keys. This coating is 

 formed with great rapidity. An exposure between two tides is 

 sufficient to form such a thin coating, as I have repeatedly had 

 occasion to observe in the deposition of finer oolitic sands which 

 fill the rock-pockets just within reach of the waves at high tide. 

 A process of undermining similar to what has been observed at 

 Key West takes place along all the coral-rock shores which 

 happen to be exposed to the action of the sea. From the de- 

 scription of Rein and others, this undermining action, operat- 

 ing on a very much larger scale on seolian deposits of consider- 

 able altitudes, must be the principal agent in the formation of 

 some of the peculiarly characteristic features of the Bermuda 

 Islands. On the east and west shore of Loop-erhead, near the 

 northern extremity, we can trace admirably the successive layers 

 of the coral limestone which have been deposited and have had 

 an opportunity to harden between the tides, forming what ap- 

 pear to be stratified beds, with their outcrops running as a gen- 



