MIGRATORY PISHES OF ASIA. 205 



Thus far it is evident that we liave fishes which breathe air, 

 die if they cannot obtain direct access to it, and whose respira- 

 tory cavities are found to contain gaseous substances, not water — 

 or proving my first proposition, that there are fishes in Asia which 

 normally respire atmospheric air direct. It now becomes neces- 

 sary to inquire what are these fishes ? and how do they differ in 

 their circulation from what obtains in ordinary fish ? 



The genera of fishes (excluding Clupeidae) which exist in the 

 fresh waters of India and possess respiratory organs having a luno-- 

 like function, and which are distinct from the gills, are as 

 follows : — 



Amongst the Acanthopterygians we have genera of the Laby- 

 rinthici, all of which possess a cavity above the upper portion of 

 the first branchial arch, which contains an elaborate apparatus 

 consisting of thin laminae of bone covered by a vascular mucous 

 membrane. Of these we have species of Anahas, Polyacanthus, 

 Osphromenus, Trichogaster. The Ophiocephalidfe have likewise a 

 cavity above and accessory to the true gill-cavity, and covered 

 with an extremely vascular mucous membrane. Amongst the 

 Siluridae, all of which possess an air-vessel in the abdomen con- 

 nected by a duct with the pharynx, we have in addition an acces- 

 sory air-breathing apparatus. The genus Clarias has a dendritic 

 accessory branchial apparatus attached to the convex surface of 

 the second, third, and fourth branchial arches ; and this is received 

 into a recess behind the situation of the normal gill-cavity ; it ap- 

 pears like sticks of red coral. From the dissections of M. Greofiroy 

 in the Nile species it would appear that these arborescent-looking 

 organizations are composed of a semitrausparent substance, their 

 external surface being densely covered by minute branches of the 

 branchial artery ; and here some at least of the blood appears to 

 be aerated. Cuvier assigns to this organ the double function of 

 affording a surface for the oxygenation of blood, and as so many 

 hearts for propelling it into the aorta. Mr. Boake (Journal 

 Ceylon Branch Royal Asiatic Soc. 1865-66, p. 133) shows that 

 this genus of fishes is one which takes in air direct. The genus 

 JSaccoiranchus has an accessory respiratory sac to the gills, which 

 extends backwards along either side of the neural spines amongst 

 the dorsal muscles of the abdominal and part of the caudal regions. 

 Amongst the eel-like forms of the family Symbranchidse, we 

 have the Cuchia eel (Amphipnous cuchia), which possesses three 

 branchial arches having rudimentary laminae and a respiratory air- 

 sac, which communicates with the gill-cavity. 



