160 THE GENESIS OF SPECIES. [CHAP. 



scribed b} T Dr. Gunther in his catalogue. 3 These forms ex- 

 tend from Java and Borneo on the one hand, to Aleppo on 

 the other. Nevertheless a new species (M. cryptacanthus) 

 has been described by the same author, 3 which is an in- 

 habitant of the Camaroon country of Western Africa. He 

 observes : " The occurrence of Indian forms on the West 

 Coast of Africa, such as Periophthalmus, Psettus, Masta- 

 oembeluSj is of the hignest interest, and an almost new fact 

 in our knowledge of the geographical distribution of fishes." 



Ophiocephalus, again, is a truly Indian genus, there 

 being no less than twenty-five species, 4 all from the fresh 

 waters of the East Indies. Yet Dr. Gunther informs me 

 that there is a species in the Upper Nile and in West 

 A frica. 



The a'canthopterygian family (Labyrinthici) contains 

 nine fresh-water genera, and these are distributed between 

 the East Indies and South and Central Africa. 



The Carp fishes (Cypronoids) are found in India, Africa, 

 and Madagascar, but there are none in South America. 



Thus existing fresh-water fishes point to an immediate 

 connection between Africa and India, harmonizing with 

 what we learn from Miocene mammalian remains. 



On the other hand, the Characinidse (a family of the 

 physostomous fishes) are found in Africa and South Amer- 

 ica, and not in India, and even its component groups are 

 so distributed, namely, the Tetragonopterina* and the 

 Hydrocyonina. 6 



Again, we have similar phenomena in that almost ex- 

 clusively fresh-water group the Siluroids. 



2 See his Catalogue of Acanthopterygian Fishes in the British Mu- 

 seum, vol. iii., p. 540. 



3 Proc. Zool. Soc., 1867, p. 102, and Ann. Mag. of Nat. Hist. vol. xx., 

 p. 110. 



4 See Catalogue, vol. iii., p. 469. 



5 Ibid., vol. v., p. 311. 6 Ibid., p. 345. 



