"288 Proceedings of the Royal Physical Society. 



inhabited by purely terrestrial animals which are not gifted 

 with means of crossing the ocean. There are, however, 

 exceptions to this rule, which are not a little puzzling — 

 such as, for example, the occurrence of Bana Guppeyi in the 

 Solomon Islands. Earthworms form another exception. 

 Apart from the islands of the Pacific, which are for the 

 most part separated from each other by such narrow tracts 

 of ocean that an accidental transfer of species is credible, we 

 have earthworms occurring in Madeira, Tenerife, St Helena, 

 Fernan [N'oronha, Marion Island, Kerguelen, and South 

 Georgia. Tenerife is included in the Palsearctic region, and 

 the justice of this conclusion is borne out by a consideration 

 of its earthworms. Through the kindness of Mr E. B. 

 Poulton, E.E.S., and Mr E. W. Headley, I have become 

 possessed of a number of earthworms from that island 

 belonging to the genera Microscolex, Allurus, and Allolo- 

 ho2:)hora. This resemblance, however, may perhaps be only 

 the result of a more active commercial intercourse with 

 Europe than with any other part of the world. St Helena 

 is inhabited by several species of earthworms belonging to 

 the genus Perichceta. But the most interesting occurrence 

 of earthworms on any oceanic islands is their occurrence in 

 Kerguelen and Marion Islands. On each of these islands one 

 species occurs, which appears to be the same. As Lankester's 

 Acanthodrilus herguelenensis was adequately described, while 

 Grube's Liimhricus kerguelarum was — to avoid all semblance 

 of exaggeration — madequately described, I retain the former 

 name, though the species are probably identical. 



There is obviously not sufficient intercourse between 

 Kerguelen and other parts of the world to account for the 

 artificial introduction of this Acanthodrilus ; and, as it differs 

 specifically from any form hitherto described, it has probably 

 occupied the islands for a considerable period. Kerguelen 

 itself is an island of considerable age, as is evinced by the 

 fact that it possesses sedimentary rocks (formed, however, 

 exclusively out of the debris of its volcanic substructure).^ I 

 point out elsewhere that Kerguelen forms part of an Antarctic 



^ Since the above was written, Mr Blanford's "Presidential Address" to 

 the Geological Society has appeared. I have quoted on p. 286 some other 



