134 J. W. SPENCKR — Rli:CONSTRUCTlON OF ANTILLEAN CONTINENT. 



Pauses in the changes of level must have taken place. Thus the 

 broad Blake plateau, at from 2,500 to 3,500 feet beneath the sea, bears 

 witness to a long period of stability ; so at the mouth and west of the 

 Mississippi river another drowned terrace at depths of 4,000 or 5,000 feet 

 repeats the same feature. The channel between Rosalind banks and Ja- 

 maica (at 3,000 or 4,000 feet), the Yucatan channel (at 7,000 feet) and the 

 submerged plateau of the Caribbean sea give concordant testimony. The 

 pauses in the elevation would explain certain fjords on the Pacific coast. 



But the writer is inclined to regard these broad terrace plains and terres- 

 trial slopes at the baselevel of erosion like the coastal plains of the conti- 

 nent, which required long periods for their completion, as representing 

 the altitude of the Pliocene continent during a considerable portion of 

 the period. In this case the Pliocene elevations were much inferior to 

 the Pleistocene altitudes, but lasted through a longer period of years. 



Following the Pleistocene elevation the depression continued so as to 

 permit the deposition of the Zapata loams, but this mid-Pleistocene sub- 

 sidence admitted the Atlantic waters to the inclosed seas, and probably 

 made narrow straits 300 feet deep between these Antillean basins and the 

 Pacific ocean, subsequently closed by the post-Zapata rise of the land. 



Thus it is seen that the Atlantic waters were admitted to the region of 

 the West Indies in the later part of the Pliocene period, to be drained off 

 by the terrestrial elevation in the earlier part of the Pleistocene days 

 with perhaps a shallow connection with the Pacific in the mid- Pleisto- 

 cene epoch, since which time there has been no connection with the Pa- 

 cific, but free communication with the Atlantic. This explains the 

 recent admission to the Antillean waters of the deep-sea type of Atlantic 

 fishes to the exclusion of Pacific forms, according to Mr Browne-Goode. 

 Of the littoral fauna of the Antillean waters only a few fishes, crustaceans 

 and mollusks (35 species out of about 1,400 according to Professor 

 Philip P. Carpenter), without the association of any echinoderms or pol- 

 yps, occur in the waters on both sides of Central America (Agassiz). The 

 Pacific species could have gained admission to the Antillean basin 

 through the straits of shallow depth that united those seas with the Pa- 

 cific in the latest Pliocene and the mid-Pleistocene depression, since 

 which time there has been no connection. 



Biologic Bearing of the physical Changes of Level. 



This study is primarily one of physical and dynamic geology. As it 

 covers such a vast region, only partly explored, errors have doubtless 

 crept in; but the general repetition and later uniformity of conditions, 

 hitherto not brought together for want of the ke}^, and the likeness to those 



