376 BULLETIN OF THE 



the marly limestone of the Oxford beds. Of course these different 

 rocks may alternate with one another, as, owing to the increase of the 

 whole formation, the conditions for the deposition of one kind of rock 

 may be followed by those favoring another combination. Afnun, in 

 consequence of the changes in the direction of the currents, or as the 

 result of a heavy gale, considerable deposits which have been goin"- on 

 regularly for a long time may suddenly be worn away and destroyed, 

 giving rise in turn to the formation of conglomerates made up of lime- 

 stone fragments of various structure, united together into very pecu- 

 liar conglomeratic pudding-stone with angular materials. The compact 

 limestones are frequently as hard as the hardest limestones of the 

 secondary formation, have a conchoidal fracture like the most compact 

 Muschelkalk of the Triasic period, and may ring under the hammer. 



Most of the keys consist of broken corals thrown up by the waves, 

 including fragments of shells, sea-urchins, and occasionally bones of sea- 

 turtles and fishes. At the Dry Tortugas and at the Marquesas, how- 

 ever, some of the keys are entirely made up of the decomposed frag- 

 ments of corallines cemented together. The crescent-shaped joints of 

 a large species of Opuntia are most prominent among them. 



Nowhere, within the range of the Gulf Stream and its borders, have 

 I -<en a rock which could be supposed to have been formed by the 

 materials accumulating in the greater depth of its trough, such as I 

 have described above, p. 3C7. And no rock in the whole Jurassic for- 

 mation could have been formed out of the kind of materials which are 

 found in the deeper parts of the Atlantic basin, along the American 

 shores ; I therefore do not believe that any of the rocks of the Jura 

 and the Suabian Alp have been deposited in very deep waters. 



The extensive area occupied by the keys and reefs of Florida, in- 

 cluding the sloping coral plateau of the American side of the Gulf Stream 

 bottom, may fairly be compared to the Jurassic formation, as it stretches 

 across Central Europe and farther east in the direction of the Caucasus 

 and Himalaya Mountains. Indeed, the Jurassic formation, as a whole, 

 bears the same relation to the older deposits upon which it rests, as the 

 modern American coral formation sustains to the older parts of the 

 coast of our continent. During the geological middle ages, the Ju- 

 rassic formation was the submarine margin of a growing continent, as 

 the Pourtales plateau forms at present the southern margin of North 

 America. 



