( 9 ) 



From these tables, whether we fix our attention on genera or on species, 

 we observe two things. 



First, that the littoral hermit-crabs of India, which include a majority 

 of specialized and, probably, lately-evolved forms, are part of a fauna which, 

 though to a certain extent showing a circumtropical distribution, is to a 

 more marked extent accumulated in the Indo-Pacific, between East Africa 

 and the eastern confines of Polynesia 



Secondly, that the sublittoral hermit-crabs of India, which include a 

 ponderous number of generalized and " primitive " forms, are pare of a fauna 

 which, though also shovvmg a circumtropical tendency, is concentrated at 

 certain definite points in the Northern Hemisphere and is correspondingly 

 deficient at certain definite points m the Southern Hemisphere. The striking 

 points of concentration are (i) the West Indian region — including, to a less 

 noticeable extent, the Bay of Panama, (2) that part of the North Atlantic 

 Ocean that includes the Azores and Cape Verde Islands, and that washes 

 the coasts of north-west Africa, and (3) Oriental Seas from Indian to the 

 Philippines (and Japan) ; and the striking deficiencies are (i) the South 

 Atlantic coasts, especially those of Africa, and (2) the south-western parts 

 of the Indian Ocean. 



If the Indo-Pacific unity of the more-recently-evolved land and littoral 

 elements of the local Paguroid fauna be a natural consequence of the open 

 communications, under uniform conditions, of the present time, then, 

 conversely, it. seems reasonable to argue that we have a clue to the ancient 

 communications of the seas of India in the concurrences of the more archaic 

 sublittoral element. 



In other words, it seems to me that the sublittoral hermit-crabs of the 

 local seas are the remnant of the sublittoral fauna of a chain of seas, or 

 archipelagos, that may, at a former time, have extended, under uniform 

 conditions, north of the equator, from Panama eastwards, by way of north- 

 west Africa, into the heart of the East Indian Archipelago 



I know that I am here sailing into very troubled waters, but it seems to 

 me that the evidence from Zoological Distribution cannot be disregarded in 

 restoring the outlines of extinct geography. 



In an Account of the Indian Deep-sea Madreporana I have specified certain 

 other elements of the Indian sublittoral fauna that have the same curious 

 residual distribution as the hermit-crabs, and have suggested an analogy with 

 the phenomenon of Alpine floras as an explanation ; and in an Accotint of 

 the Deep-sea Brachyura collected by the ''Investigator" I have drawn attention 

 to certain sublittoral genera of crabs that exhibit the same peculiarity of 

 range. 



