RELATIONSHIP OF THE FAUNA. 85 



have been of mammals, reptiles, and amphibians, leaving but 

 a few groups bats, rodents as possible survivals. Nor could 

 we reasonably expect to find the remains of such animals 

 preserved either in the coral rock or in the drift-rock of the 

 islands. With regard to the question of the recent origination 

 of the islands the evidence from the vertebrate fauna proves 

 little. The great distance of the islands from the mainland in 

 itself explains the poverty of the fauna, whether this be old or 

 new, while the absence of distinctive or special types among 

 birds is, as Mr. Wallace well holds, due to the too frequent 

 crossing of migrants or involuntary wanderers, which keeps 

 the various breeds true, and prevents specific modification. 

 The absence of peculiar species is, therefore, not a result of 

 newness ; on the contrary, certain considerations seem to indi- 

 cate that the island group is of greater antiquity than has 

 been generally assumed, and not impossibly some of the lower 

 forms of life now inhabiting it are descendants of an ancient 

 fauna which was well developed before the present physical 

 conditions were established. 



An analysis of the Bermudian invertebrate fauna shows 

 some very interesting and remarkable features, which prove 

 the complexity of zoogeographical inquiry. Before our visit 

 but little systematic work for most of which we are indebted 

 to Matthew Jones was done in this department of zoology, 

 and, doubtless, much still remains to be done. But the general 

 features of this fauna are now sufficiently determined to per- 

 mit of satisfactory conclusions being drawn from them. 



The marine Mollusca of the archipelago, which up to the 

 time of our visit were listed at about 80 species, comprise, as 

 far as is now known, some 170 species. These, with probably 

 less than a dozen exceptions, are all members of the West 

 Indian or Floridian faunas. Lying in the path of the Gulf 

 Stream drift, which strands upon the island-shores vast quanti- 

 ties of the Gulf-weed (Sargassum bacciferum), there seems little 

 reason to doubt that by far the greater number of these forms 



