104 PROCKEDINGS OF THE 3IALAC0L0GICAL SOCIETY. 



extreme uoitli and south limits of our own P. lapillus, L., no less in 

 Europe and Africa than on both sides of North America and in Jajian, 

 and whether Adanson was right or wrong when he enumerated it 

 among tlie ^lollusca of Senegal.^ 



Purpura hcBmastonia, L., has long been known to inhabit both sides 

 of the tropical Atlantic, from West Africa, the Mediterranean, and 

 the coasts of Portugal and France on the one hand, to Erazil, the 

 AVest Indies, and the southern states of Eastern jS^orth America on 

 the other. But it is only of recent years that we have learned that 

 P. coronata, Lam., has crossed the Atlantic too, and has appeared in 

 Demerara^ and Trinidad and in East Guatemala.^ The West Indies 

 liave retaliated by sending to AVest Africa a form [P. eudeli, Sow.) 

 whose rehitionship to P. patida, L., is so close as to leave no doubt of 

 its origin, and scarcely any that it should be counted as a mere 

 variety.'' Has this process of exchange between the shores of the 

 Atlantic at its narrowest part, over 1,600 miles, gone any further, 

 e.g. in any form of Littorina possessing a free-swimming larva? 

 The transit of the larval form from one coast to another would be 

 facilitated by the remarkably equable temperature of the intervening- 

 water (a steady 77°-80° F. all the year round), by the absence of 

 any strong north or south current, and by the more or less circulatory 

 drift of water between the two continents. 



If larval forms of Purpura can pass from AVest Africa to South 

 America, and vice versa, it is easy to understand how P. columellaris, 

 Lara., an obvious derivative of P. patula, L., became established at 

 the Galapagos, only 600 miles from the nearest mainland. The 

 heated water of the Bay of Panama follows the coast southward until 

 it reaches Cape San Lorenzo, in lat. 1° S., where it is deflected 

 westward, straight for the islands. Trees from the mainland, with 

 the leaves still upon them, have been found cast up on the island 

 shores. The molluscan fauna of the Galapagos thus exhibits large 

 contributions from the Panamic and Peruvian regions, with a very 

 slight admixture of the Indo-Pacific element.* 



Again, P. neritoidea, L., is a common West African littoral shell. 

 It is also found in the Cape Verde, 300 miles from the coast, and, as 

 a variety scarcely distinguishable from the type, on Ascension Island, 

 nearly 900 miles from the nearest African land. 



Further research on the relationships of adjacent groups of Purpura 

 would probably bring out valuable results, for the genus is almost 

 worldwide and abundant in species and in individuals. Some light 

 might be thrown on the remarkable way in which it is replaced, on 

 the coasts of Chili and of the ilagellanic and part of the Patagonian 



^ M. Adanson, Hist. nat. Senegal : Coquillages ; Paris, 1757, pp. 106-7, 



pi. vii, tig. 4. 

 - A. H. Cooke, Journ. Malac, vol. iv, p. 69, 1895. 

 •• H. A. Piisbry, Nautihis, vol. xiii, p. 130, 1900. 



■* The species was described by Sowerby in Journ. Conch., voL x, p. 74, 1903. 

 ^ These facts are due to W. H. Dall, Report on a Collection of Shells from 



Peru, etc. : Smiths. Inst. Proc. U.S. Nat. Mus., vol. xxxvii, pp. 147-294, 



1909. 



