232 ANNUAL EEPORT SMITHSONIAN INSTITUTION, 1915. 



From these two primary facts M. Germain deduces the evident 

 conclusion that the four archipelagoes were connected with the 

 African Continent up to an epoch very near .our own, at the very 

 least until toward the end of the Tertiary. 



Thirdly, in the present Mollusca of the four archipelagoes there 

 are some species which seem to be the survivors of the fossil species 

 of the European Tertiary; and a similar survival exists also in the 

 vegetable series, a fern, the Adianfuni reniforme^ at present extinct 

 in Europe, but known in the Pliocene of Portugal, continuing to-day 

 to live in the Canaries and in the Azores. 



M. Germain deduces from this third fact the bond, up to Pliocene 

 times, with the Iberian Peninsula, of the continent which embraced 

 the archipelagoes and the severing of this bond during the Pliocene. 



Fourthly, the Pulmonata Mollusca, called Oleacinida?, have a pecu- 

 liar geographic distribution. They live only in Central America, 

 the West Indies, the Mediterranean Basin, and the Canaries, Ma- 

 deira, and the Azores. In America they have preserved the large 

 size that they had in Europe in the Miocene epoch ; in the Mediter- 

 ranean Basin and in the Atlantic islands they have become much 

 smaller. 



This geographic distribution of the Oleacinidse evidently implies 

 the extension to the West Indies at the beginning of the Miocene 

 of the continent which included the Azores, the Canaries, and 

 Madeira, and the establishing during the Miocene, or toward its close, 

 of a separation between the West Indies and this continent. 



Two facts remain relative to the marine animals, and both seem 

 impossible of explanation, except by the persistence, up to very 

 near the present times, of a maritime shore extending from the West 

 Indies to Senegal, and even binding together Florida, the Bermudas, 

 and the bottom of the Gulf of Guinea. Fifteen species of marine 

 Mollusca lived at the same time, both in the West Indies and on the 

 coast of Senegal and nowhere else, unless this coexistence can be 

 explained by the transportation of the embryos. On the other hand, 

 the Madreporaria fauna of the island of St. Thomas, studied by M. 

 Gravier, includes six species — one does not live outside of St. 

 Thomas, except in the Florida Eeefs; and four others are known 

 onl.y from the Bermudas. As the duration of the pelagic life of the 

 larva? of the Madreporaria is only a few days, it is impossible to 

 attribute this surprising reappearance to the action of marine 

 currents. 



In taking all this into account, M. Germain is led to admit the 

 existence of an Atlantic continent connected with the Iberian Penin- 

 sula and with Mauritania and prolonging itself rather far toward the 

 south so as to include some regions of desert climate. During the 

 Miocene again this continent extends as far as the West Indies. 



