1871.] FRESHWATER SILUROIDS OF INDIA. 717 



Panyasius, Pseudeutropius, Callichrous, Wallayo, Olyra, Silurus, 

 Plotosus, whilst Chaca is at present doubtful. The general form 

 of the air-vessels is elongated, or heart-shaped, some being poste- 

 riorly prolonged. Internally there is an internal septum dividing 

 the organ into two lateral halves, which are sometimes further sub- 

 divided by transverse partitions. The central longitudinal septum 

 has an orifice anteriorly affording communication with the two 

 sides ; whilst all the lateral chambers communicate with one another 

 on the same, and by means of the anterior one with those on the 

 opposite side. 



Of those with the air-vessel wholly or partially enclosed in bone, 

 we have Clarias, Saccobranchus, Silundia, Ailia, Ailiichthys, Eutro- 

 piichthys, Sisor, Gayata, Hemipimelodus, Bagarius, Pseudecheneis, 

 Glyptosternum, Amblyceps, and Exostoma. The general form of 

 these air-vessels is transverse, with the outer extremities usually 

 dilated and enclosed in bone. Or the air-vessel may be in the form 

 of two rounded lobes enclosed in bone, with a connecting tube, or 

 else the pneumatic ducts coalesce a short distance before they enter 

 the pharynx or upper portion of the alimentary canal. 



This last division has much in it that is Loach-like in its smaller 

 forms ; and Mr. Blyth's observation of Amblyceps, a " Cobitis-Mke 

 Siluroid," is still further apparent when we examine its air-vessel. 

 Thus it gradually leads the Siluroids towards the Cobitidina. 



I do not propose at present to enter further upon the subdivision 

 of the Siluroids, as I hope those who have the opportunity of giving 

 descriptions of the air-vessels of extra- Indian genera will do so, in 

 order to ascertain whether this division is a natural one and appli- 

 cable to the whole of the family*, as I anticipate it will be found 

 to be. 



The following notes upon the geographical distribution of these 

 fishes are far from complete ; but I deem it advisable to give them 

 as they are, for were I to wait until all that is desirable is effected, I 

 fear the paper would never be finished. 



Previously to commencing the geographical distribution of the Si- 

 luridce, I have some remarks to offer respecting the limits chosen 

 and the terms employed. 



By India or Hindoostan I understand the whole of the continent 

 under British rule or protection, or independent native states when 

 surrounded by British territory. Burtnah includes Arracan and the 

 Tenasserim provinces, whilst the designation Ceylon answers for 

 itself. 



The East-Indian archipelayo has been commonly employed in 

 zoology to express in a general way a large space variously limited 

 or expanded by each fresh author, aud is scarcely sufficiently defined 

 when entering upon the distribution of freshwater genera or species 

 offish. I shall therefore employ Wallace's definition of the Malay 



* Any naturalist who wishes to exchange Tropical forms of Siluroids, person- 

 ally collected, for those from Hindoostan, will always find me ready to meet his 

 views. The reason why I wish them personally collected is that the localities 

 may be correctly appended. 



