626 PATAGONIAN EXPEDITIONS I ZOOLOGY. 



lommatinidce (Adelopoma], Clausiliidce, etc. These forms never have the 

 old-age stigmata of the preceding group. They are developed in won- 

 derful abundance and virility. Being known in characteristic genera of 

 American type (Pleurodonte, Cepolis] in the Floridian Oligocene island, the 

 advent of the group in middle America must have been much earlier. It 

 could hardly have been later than the beginning of the Eocene, and prob- 

 ably was not later than the Upper Cretaceous. 1 



Finally, we have as the latest faunal element in the Antillean-Mexican 

 area, a series of South American forms Achatinidce, Bulimulidce, 

 Ampullariidce, Melaniidcz, Mutelidce and some North American forms, 

 Unionidce, Pupillida, Zonitidce, Polygyrince, etc. These are, with very few 

 exceptions, unchanged generically, and some are specifically identical with 

 existing South or North American forms. It is very evident that such 

 Archhelenic forms as exist in the Antilles and Mexico were not derived 

 directly from the Archhelenic area ; they migrated in the later Tertiary 

 and Pleistocene from the Guiana-Colombian center. 



Antillia has not been an evolution center for fresh-water mollusks or 

 fishes, evidently because it has never been a very large area, and has been 

 an unstable one, at one time in form of a continent, again an archipelago, 

 hence without river systems of great extent or duration, such as are essen- 

 tial to the evolution of a fresh-water fauna. 



There is absolutely nothing in the distribution of mollusks suggesting 

 that either South or North America was at any time connected with the 

 supposed South Pacific continent, or the Hawaiian group. Even Juan 

 Fernandez has a land-snail fauna of Pacific and not South American type. 

 The Hawaiian and Polynesian connections with America mapped by Arldt 

 (1907) seem quite impossible. 



1 The American Clausiliidcz are thought by Professor Boettger to be related to the European 

 Miocene and recent Pyrenean group Laminifera, but they seem to me even closer to the genus 

 Garnieria, of the Indo-Chinese center. The Helicinidce may have arisen in Middle America from 

 an aquatic rhipidoglossate stock, but the very wide distribution of the group in Polynesia and 

 eastern Asia suggests that it is a very old one, which probably appeared among the first land- 

 snails. 



The Belogonous Helices still exist in high latitudes on both sides of the Pacific, being known 

 from Sitka on the American and the Kuril Islands on the Asiatic side. 



