286 Proceedings of the Roijai Irish Acadein>j. 



And he contended that the West Indian ten-estrial moUuscan fauna 

 had beeniDartly derived from the IMiocene fauna of Europe (B, p. 147- 

 148). He is no^r convinced that the Atlantic Islands must have been 

 joined to the mainland in lliocene times (D, p. 53) ; though he does not 

 comment upon the question whether the islands have had a continental 

 connection since. He still leaves it to be explained how the strong 

 Mediterranean element reached the islands. Dr. von Ihering solves 

 this problem in a peremptory manner by saying that no malacologist 

 nowadavs could explain the presence of these continental Molluscs on 

 the islands in any other way but by their progression on land 

 (A, p. 51). 



The moUuscan faima of the Atlantic Islands does not apparently 

 lend any support to the theory that they formed part of an ancient 

 land-bridge joiaing oiu" Continent to the "West Indies or ^S'orth 

 America. Eut a study of the Mediterranean fauna reveals a certain 

 relationship with that of the "\\^est Indies in the genera Glandina, 

 Tudora, and Leonia. This resemblance of the two faunas, liowever, 

 becomes much more marked when we compare the extinct land 

 Molluscs of Europe with those at present living on the other side of 

 the Atlantic. A migration from our Continent across Asia and Bering 

 Strait to America, in explanation of these facts, is inadmissible, as the 

 species in question have never been found fossil in either of the last 

 two continents. A direct land-connection between Europe and America 

 must therefore have existed across the Atlantic in Miocene times, 

 the Molluscs migrating fi'om the foiiner to the latter. Dr. Kobelt 

 places this land-bridge much fiulher north than the Atlantic Islands 

 (B, p. 147). I should, on the contrary, be inclined to locate it to the 

 south of these islands, as it seems evident that Glandina, Tudora, and 

 Leonia have only spread from the "West Indies into ]Srorth America 

 proper in comparatively recent times. The fact that these species, 

 which have representatives in the "VTest Indies, are more or less 

 confined to the Mediterranean region in Eru-ope, also points t> a 

 southern connection rather than a northern one; and it is quite possible 

 that the migration took place along the northern coasts of the land, 

 which is supposed to have united Af lica and South A merica, and in 

 favour of which Mr. Murray, Dr. Blanford, and Dr. von Ihering 

 especially have brought forward many important distributional 

 evidences. 



Dr. von Ihering produces evidence of a strong resemblance 

 between the Brazilian and the West African invertebrate coast fauna. 

 He al"50 draws attention to the fact that the fresh-water mussels of South 



