CHANGES OF LEVEL OF NOllTH AMERICAN COASTS. 155 



in the limestone rock with all their parts ahove the erosion hasclevol, we 

 liave to supi)ose a considerable subsidence to account for these inverted 

 siphons through which the rainwater courses to the sea. I am not aware 

 that any soundings have been made which serve to show the depth of 

 the sea innnediately over these submerged cavern mouths. It is, indeed, 

 not likely that such soundings would give evidence of value as to the 

 original horizontal plane of the exits, for under the existing conditions 

 the streams would tend to cut away the roofs of the caverns near the 

 mouths and to till in the fioors at those points with debris, so that the 

 exits may have worked upwardly for an unknown height. Therefore 

 these submarine si)rings, as with like i)henomena along otlier coastlines, 

 while indicating the downsinking of the land, afford no useful gauge as 

 to its amount. 



There is yet another set of facts which serves to show that the penin- 

 sula of Florida has recently stood at a much greater height than it now 

 occupies. The wells which have been l)ore(l in the district, some of 

 which have attained the depth of more than 1,000 feet, show that the 

 water of deposition — that is, that of the seas in which the marine strata 

 were formed — has, to the depth of 800 feet or more, been almost alto- 

 gether displaced by rainwater. It does not seem to me possible to ac- 

 count for this leaching out of the construction water except on the sup- 

 position that the area has recently been much higher than it is at present. 

 It is true that the time of this elevation cannot be determined, but the 

 age of the strata affected seems to indicate that the event occurred in com- 

 paratively recent Tertiary time. 



It seems to me that the three groups of evidence above noted clearly 

 establish the fact that along the Floridian portion of the coastline there 

 has been a recent subsidence, the depth of which is to be measured by 

 hundreds of feet. 



Florida to Delaivare Bay. — North of Florida and thence to the southern 

 limits of the ice-sheet of the last glacial epoch the evidence from flooded 

 valleys, though as far as Hatteras it is more or less masked by barrier- 

 beaches and swamp accumulations, is unmistakably clear. None of the 

 rivers exhibit deltas, except slight accumulations at the head of their 

 reentrants. All of them are flooded for a considerable distance from the 

 open sea. In the Dismal swamp district, moreover, as I have noted in 

 a report on that district,* some of the streams within the morass have 

 their existing bottoms much below the level of the neighboring sea-fioor. 



Delaware River. — Delaware river is the last stream the mouth of which 

 is south of the glacial field. It is therefore the most northern of the 

 Atlantic coast rivers where the effects of flooding due to subsidence are 



* Annual Report of the United States Geological Survey, 1890, p. 320. 



