428 WEST INDIAN MOLLUSES— SIMPSON. vol. xvir. 



and altliougli nearly half tlie species of land snails of these four larger 

 islands are opercnlates, not more than 9 or 10 of them are found outside 

 of a single island. This might be taken as evidence that the non-oper- 

 culate forms were of much more ancient origin than the others, many 

 of them reaching back to a time of former land connection, while the 

 operculate species were of more recent development, which I suspect 

 mav be the case; or that the former are better adapted to migration 

 across the ocean than the latter. 



The fact that the opercnlates form so large a proportion of the 

 Antillean land snail fauna, that a majority of the genera are fouiid on 

 two or more of the islands and the mainland, while nearly every species 

 is absolutely restricted to a single island appears to me to be very strong 

 testimony in favor of a former general land connection. 



I believe that all the evidence of the terrestrial and tiuviatile mol- 

 luscan fauna of this region indicates that in the early Tertiary Period, 

 perhaps, there was a general land elevation of the Greater Antilles, and 

 possibly of some of the adjacent area; that Wallace's theory of a land 

 connection of the greater islands is correct; that during some part of this 

 time a landway extended across to the continent;* that the species and 

 groups of this then connected territory migrated to some extent from 

 one part of it to another, and that a probable connection existed over 

 the Bahama plateau to what was at that time no doubt the island of 

 Florida. It would appear that at this time the volcanic islands of the 

 Lesser Antillean chain were not yet raised above the sea, or that if 

 there was land in that region it has. since been submerged, and there 

 seems to be no good evidence in favor of any land connection with the 

 Greater Antilles since the lifetime of the present groups and species 

 of West Indian land and fresh-water mollusks. 



We have not as yet a sufficient knowledge of the geology and palae- 

 ontology of this general region, or a large enough acquaintance with 

 the distribution of the terrestrial and tiuviatile mollusks in Central and 

 South America to at all fully trace the past history of the region, or of 

 the forms of life in question, and, therefore, most of these theories and 

 speculations are advanced with the utmost- caution, and rather as sug- 

 gestions, subject to modification by future discoveries, than as absolute 

 explanations of the facts. Yet enough is known to make many points 

 reasonably certain. 



Bland has divided the Greater Antilles into five different sub-prov- 

 inces: t(l) Cuba, with the Isle of Pines, the Bahamas, and Bermuda; 



* It is quite probable that at this time Mexico and most of Central America formed 

 an island ; that the sea flowed through what is now the Isthmus of Panama; and 

 that there was connection by a strait from the Gulf of Mexico through or around to 

 the northern end of the Gulf of California. In using the expression "a landway 

 across to the continent" I mean to what is the continent now. 



t " Geographical distribution of the West India land shells," Am. Lye. Nat. 

 Hist, of N. Y., VII., p. 346. 



