398 H- B POLLARD, 



superior and obliquus superior, over the eyestalk. In TricJiomycierus 

 it runs on the outer side of the olfactory organ to reach the tentacle. 

 In Clarias the nerve does not bear quite the same relation to the 

 eye muscles, because the small eye, along with the muscles, is shifted 

 very far laterally, while the nerve follows the skull wall more closely. 

 It runs on the median side of the olfactory organ to reach the ten- 

 tacle. In other Siluroids the ophthalmicus profundus may only be 

 represented by some fibres running along with tlie ophthalmicus super- 

 ficialis of the Facial. 



In some Siluroids e. g. Silurus (Stannius) and Clarias the 

 trochlear nerve runs along with the ophthalmicus profundus. That 

 is purely a case of apposition. 



The comparative anatomy of the ophthalmicus profundus is fairly 

 well known in the vertebrates, and need not be further entered into 

 here, except as regards Myxine. In Cyclostomi, Müller and Stannius 

 showed that it contains motor fibres, a fact of which most embryo- 

 logists who have drawn up schemes of the head have been oblivious. 

 The motor fibres supply muscles attached to and working the nasal 

 tube and belonging to the tentacular system. Another feature in 

 Myxine is that a small twig from the ophthalmicus profundus supplies 

 the premaxillary tentacle. 



Premaxillary Tentacle. 



A premaxillary tentacle occurs in Cobitidae (Misgurnus) and 

 some Cyprinoids e. g. Barhus, where it is attached to the premaxilla. 

 However, these forms are much modified as to histology and topo- 

 graphical relations and are not suited to form a basis for comparison. 



In Siluroids, though the tentacle itself is absent (except in 

 Aspredinidae Günther), yet there is usually present a block of pro- 

 cartilage at the tip of the snout, in relation with the premaxilla. In 

 Callichthys (Figs. 6 and 7 Pmx.p.) this block is triradiate, occupying 

 the tip of the snout below the prepalatiue pieces, and on its lower 

 surface develops the premaxilla. A corresponding block is present in 

 Chaetostomus, differing somewhat in shape. With it articulate the 

 anterior ends of the preniaxillae whicli possess the remarkable form 

 depicted by several authors in the Hypostomidae. 



The ventral position of the mouth in these South American Silu- 

 roids is due to the large size of this piece and of the prepalatine, in 

 addition to the jjresence of a rostrum. In Sturgeons it is due mainly 

 to the rostrum, while in Selachii it is brought about by the position 



