140 J.W.SPENCER — RECONSTRUCTION OP ANTILLEAN CONTINENT. 



region. Accordingly it is concluded that the Antillean bridge stood 

 at from one and a half to two and a half miles above the present alti- 

 tudes of the plains that now form the islands, with their mountains rela- 

 tively somewhat lower than at present. The floors of the Mexican gulf 

 and the Caribbean sea were then plains, and there seems to be strong 

 reason for inferring that these plains drained into the Pacific ocean. 



The formations out of whicli the valleys are excavated belong mostly 

 to the more recent geologic periods, and are generally but little dis- 

 turbed. From the determination of their age and that of the materials 

 filling the buried valleys, it has been found that there have been two 

 epochs of great elevation, namely, in the Pliocene and in the Pleistocene 

 periods. . Between these there was a subsidence of such depth as to 

 drown the continental coastal plains and reduce the West Indian region 

 to very small islands, with (probably) a shallow connection between the 

 Atlantic and the Pacific oceans. The mid-Pleistocene depression was 

 not quite so great as the earlier, and there was probably a strait connect- 

 ing the two oceans. Since that time there have been several oscillations 

 of minor degree, with the formation of many small coastal canyons and 

 the elevation of terraces and coral reefs. 



The Central American barrier did not sink to the depths of the Antil- 

 lean seas in the great subsidence, but this has been further raised in 

 part by the epeirogenic movements and thrown into a great barrier by 

 orogenic movements and volcanic action. 



The Neocene and Pleistocene mammalian fauna suffered from both 

 the changes of climate, passing from the tropical to the almost arctic, and 

 the consequent changes in the food supply, and the extermination was 

 completed by the depression of the broad areas below the sea. 



This study establishes the great mobility of the earth's crust, and the 

 application of the methods reaches far beyond the region investigated in 

 this contribution and opens many new problems in dynamic geology. 



Note (see page 118).— A study of the slopes of the drowned valleys might have been desirable, 

 but it should be noted that their mean slope could not have been compared with that of existing 

 rivers except those of mountain plateaus, for most of the valleys of the eastern part of the conti- 

 nent are buried in their lower reaches by recent accumnlations, and therefore do not represent 

 their true slope of erosion. In the denudation of the high tablelands, even when composed of 

 loose materials, the valleys deepen slowly, the great excavations being made at the points of sud- 

 den descent. Here the deep valleys are rapidly elongated, their length partly depending upon 

 their age.. The analogy between the slopes of the drowned valleys and that of the valley of the 

 Colorado river is close, but comparison could not properly be made with one like that of the Mis- 

 sissippi river. 



