362 CORALS AND CORAL ISLANDS. 



These are examples of the comparative!} 7 rapid formation 

 of caverns. The waters which run or percolate through them 

 must be charged with carbonic acid to accomplish such work, 

 and yet they have no source for this ingredient except the atmos- 

 phere, animal respiration, and vegetable and animal decom- 

 position in the soil. The flutings and stalactitic incrustations 

 of a precipice facing the sea must depend on the former alone, 

 with the aid perhaps of the spray from the sea brought over 

 the reef by storms. 



XIII. OCEANIC TEMPERATURE. 



Facts seem to indicate — though perhaps not sufficient to 

 demonstrate — that the Gulf Stream has had, from the Juras- 

 sic period in Geological history onward, the same kind of in- 

 fluence on the temperature of the North Atlantic Ocean which 

 it now has. 



The existence of a coral reef made out of corals of the As- 

 trsea tribe and others, during part of the Oolitic era (middle 

 Jurassic), in England, as far north as the parallel of 52° to 

 55° is strong evidence that the isocryme of 68° F., the coral- 

 reef boundary, extended then even to that high latitude ; for 

 species of the Astrsea tribe are now confined to coral-reef seas 

 (p. 109). This isocryme now reaches along the course of the 

 Gulf Stream, to a point just north of the Bermudas, near 

 33° N. ; and 55° is 22° beyond this. 



There are no marine fossils in any rocks of that period on 

 the American side of the Atlantic, so that facts fail for clefi' 

 nitely locating the western terminus of this oolitic isocryme of 

 68° F. But it is highly improbable that the whole ocean 

 across, on, or near, the parallel of 55° N., should have had, as the 

 mean temperature for the coldest month of the year, one so high 

 as 68° F. ; the present average position of the isocryme of 68°F., 



