X BARRIERS TO DISPERSAL 28 1 



stretches, with scarcely a break, from the west coast of Africa 

 to the extreme east coast of Asia. The Mediterranean offers no 

 effectual barrier ; shells of southern Europe are found in pro- 

 fusion in Morocco, Tunis, and Egypt, while all through Siberia 

 to the extreme of Kamschatka the same types, and even the 

 same species, of Mollusca occur. 



A detailed examination of the means, other than voluntary, 

 by which Mollusca are transported from one place to another 

 hardly comes within the scope of this work. Ocean currents, 

 rivers, floods, cyclonic storms of wind, birds, and even beetles 

 and frogs, play a part, more or less considerable, in carrying 

 living Mollusca or their ova, either separately or in connexion 

 with floating debris of every kind, to a distance from their 

 native home. Accidental locomotion, of one or other of these 

 kinds, combined with the well-known tenacity of life in many 

 species (p. 37), may have contributed to enlarge the area of dis- 

 tribution in many cases, especially in the tropics, where the 

 foi'ces of nature are more vigorous than in our latitudes. The 

 ease with which species are accidentally spread b}" man increases 

 the probability of such cases occurring without the intervention 

 of human agency, and numbers of instances may be collected of 

 their actual occurrence.^ 



A point, however, which more concerns us here is to remark 

 on the exceedingl}" wide distribution of the prevailing forms of 

 fresh-water Mollusca. It might have been expected that the 

 area of distribution in the fresh-water forms would be greatly 

 restricted, since they cannot migrate across the land from one 

 piece of w^ater to another, and since the barriers between pond 

 and pond, lake and lake, and one river system and another are, 

 as far as they are concerned, all but insuperable. We might 

 have expected, therefore, as Darwin and Wallace have remarked, 

 to find a great multiplicity of species confined to very restricted 

 areas, since the possibility of communication with the parent 

 stock appears, in any given case, to be so exceedingl}^ remote. 



As is well known, the exact reverse occurs. The range, not 

 merely of genera, but even of individual species, is astonishingly 

 wide. This is especially the case with regard to the Pulmonata 

 and Pelecypoda. The genera Limnaea, Planorhis^ Physa, An- 

 cylus^ Unio^ and Cydas are world-wide. Out of about ten genera 

 1 Mr. H. W. KeAv, TJie Dispersal of Shells, has brought together a very large series. 



