ScHARFF — Some Remarks on the Atlanfis ProUem. J?9o 



for careful consideration in relation to the theories of past changes of 

 land (p. 57). 



As regards the earthworms of the Atlantic Islands, most of them 

 are identical with European or North African forms. "Whenever 

 peculiar species occur, — such as Helodrilus Mohii oi Madeira — they are 

 closely allied to continental ones. The particular species mentioned 

 has its nearest relation {H. Molleri) in Portugal. Several of the rarer 

 forms, — such as Lendrolcena madeirensis of Madeira and Bimastus 

 Mseni of the Azores, — occur also on the part of the Continent nearest to 

 them — viz., in Portugal. 



As for any indications of an American relationship among the 

 Atlantic Island earthworms, there are some species — such as Fixer etima 

 californica of Madeira and P. bariadensis of Teneriffe — which may 

 be of American origin ; but they probably owe their existence on the 

 islands to artificial introduction. 



The problem of the former land-connection between America and 

 Africa also receives some support from the distribution of earth- 

 worms. The family Geoscolicidae is almost entirely confined to 

 South America and Africa ; only a few species reach the Palsearctic and 

 Oriental regions. 



Among the genera which indicate the former union between the 

 two continents might be mentioned Gordiodrilus — which is confined to 

 the West Indies, the Gold Coast, and Zanzibar — and Nematogenia, 

 which is only found at Lagos in West Africa, and at Panama in 

 Central America (cf. Michaelsen). 



Yery little is, as yet, known of the leech-fauna of the Atlantic 

 Islands, but our common European horse-leech {Mamopis sanguisuga) 

 has been met with in the Azores ; while Di7ia Blaisei, a common 

 Mediterranean species, occurs on several islands of the Azorean 

 archipelago, and also on Madeira. 



Conclusions. 



The conclusions based upon this general siuwey of the fauna of 

 the Atlantic Islands are of more value than if only a single group 

 of animals had been taken into consideration. It will also be con- 

 ceded that, fi'om the facts and examples I have collected, we are 

 entitled to form a very definite judgment on the subject, though my 

 interpretation of these may not appeal to all. 



The great importance of such data of distribution, as factors in 

 solving problems of former geographical changes, is now generally 

 recognized. Dr. Wallace was the first to appreciate the bearing of a 



R.I.A. PROC, VOL. XXIV., SEC, 13.] 2 B 



