DISTRIBUTION OF INDIAN FRESHWATEB PISHES. 577 



rations from its dorsal spine ; and in time, if this continues, it 

 will become more like B. terio, H. B., than the typical form. 



On the other hand, some Carps would appear to be more derived 

 from the east, as : — Chela, from the Malay archipelago to the Hin- 

 dustan and Ceylonese subregions ; Semiplotus, from Burma to 

 Assam ; Catla, also from Siam and Burma to the Hindustan (but 

 not the Ceylonese) subregions ; AmblypJiaryngodon, Danio^ Peri- 

 lampus, Nuria, and ItoTitee, from Burma to India generally. 



If we now turn to the Siluridge or scaleless fishes, we find them 

 represented by 26 genera composed of 85 species, demonstrating 

 how inferior in numbers they are to the Carps. Some of these 

 genera, as Clarias and Saccobranchus, have, as I formerly ob- 

 served (" On Amphibious and Migratory Fisbes of Asia," Journ. 

 Linn. Soc. Zool. vol. xiii. p. 198 ef seq.), respiratory organs having 

 a lung-like function, and which are distinct from the gills ; and 

 as all fishes having these accessory breathing- organs are restricted 

 to tropical regions, we may assume that Clarias and Saccolran- 

 cJiiis are tropical fishes. 



I have already (Journal Linn. Soc. Zool. xii. p. 338) given an 

 account of these fishes as found in India with the localities they 

 inhabit ; and it is therefore unnecessary to adduce further rea- 

 sons for considering that we have the remains of three distinct 

 and separate faunas existing amongst the living freshwater 

 fishes of India. The Jirst, among the ancient granitic hills 

 of the Western Grhauts, extending into Ceylon, and also found 

 on the Himalayas and in the Malay archipelago, shows some 

 former connexion between these various points. That the fish 

 themselves are of two races — the Palsearctic, which were derived 

 from Asia (or the Mediterraneo- Persic subregion) west of 

 the Indus ; and the Malayan, which came through a continent 

 now submerged beneath the Indian Ocean, a portion of which 

 we, however, still discern in the Andamans and Nicobars. The 

 second fauna, that of the plains, has an African element in it, 

 and was likewise derived by a land communication west of the 

 river Indus ; but, due to some cause, its genera, unless widely dis- 

 tributed, give but a small proportion of existing forms. The 

 third fauna, and by far the largest, is spread over the plains of 

 India, and derived its existence through communication being 

 formed with Burma and countries to the eastward ; and these ap- 

 pear to have supplanted the prior African element from the waters 

 of the plains. 



LINN JOUBN. — ZOOLOQT, VOL, XIV. 42 



