Certain genera and species of Crabs ; so that the evidence to be adduced from the 

 fish-fauna will already have been corroborated from several other marine groups. 



It is no part of my plan, in this series of Reports, to refer to the evidence 

 that lies outside the boundaries of the " Investigator" collections ; but no one 

 who remembers the singular distribution of certain Mammals (e.g. the Tapirs), 

 Birds (e.g. the Trogons, Barbets, and Whistling Teal), and Amphibia (e.g. 

 the Cmciliidse), and no one, more especially, who has considered the geogra- 

 phical range of a number of Indian Snake and Lizard families and genera 

 too numerous to mention, can have failed to have suspicions of the former ex- 

 istence of some land connexion (which this sea-connexion implies) between the 

 tropical and subtropical regions of America and the Old World — suspicions 

 which the comfortable formula " similarity of conditions" alone will not satisfy. 



Moreover, as regards marine fishes, the researches of Dr. Giinther have 

 familiarized us with the idea of a former direct and open connexion between the 

 Mediterranean and Japanese Seas (Introcl. Study of Fishes, p>- 270). 



The testimony supplied by the Indian fish-fauna favourable to the hypo- 

 theses in question may be briefly summarized as follows : — 



A. The Marine Fishes. The total number of genera of Indian marine 

 fishes is about 350, of which 



(1) about 28 per cent, are common to the Atlantic, the Mediterranean, and 

 the Indo-Pacifie : 



(2) nearly 6 per cent, are common to the Atlantic, the Bed Sea, and the 

 Indo-Pacifie, to the exclusion of the Mediterranean : 



(3) nearly 20 per cent, are common to the Atlantic and Indo-Pacifie, to the 

 exclusion of the Mediterranean and Red Seas : 



(4) nearly 3 per cent, are common to the Mediterranean and Indo-Pacifie 

 to the exclusion of the Atlantic. 



That is to say, taking the Atlantic and Mediterranean as one integral region, 

 over 56 per cent, of the genera of Indian marine fishes are also found in that 

 region. 



Not only so, but a considerable number of species are common to the two 

 regions, and although it may be argued that most of these — such as the Sword- 

 fishes, the Pilot fish, the Tunnies, the Sucker fish, the Stemoptychidse, and even 

 certain Glupeidss and Beryeidx — are oceanic forms of unlimited range, yet this 

 objection cannot, I think, apply to such species as Lobotes surinamensis, Bembrops 

 caudimacula, Chaunax pictus, or Macrurus cavernosus, or to Sargus noct and 

 Crenidens Forslcalii. 



B. The Freshwater Fishes. Exclusive of certain immigrants from marine 

 families, the freshwater fishes of India are Carps (which are nearer two-thirds 

 than half the whole freshwater fish-fauna), Siluridse, Cyprinodontidse, Ghromides 



