358 



Amphioxusliko ancestor. Adopting a term of Owen's 

 for the group of which Amphioxus is a member I have 

 designated this theory as the „cirrhostomial" theory. 



It is often stated that there is a sharp distinction between the 

 Cyclostomi and Gnathostomi in the roundness of the mouth. It is 

 however a wellknown fact that the roundness of the mouth in the 

 lamprey has nothing to do with the roundness of the mouth in Myxin- 

 oids but it is not quite so well known that many of the Siluroids 

 can scarcely be said to have a biting mouth at all. Thus in Chaetos- 

 tomus the long teeth which resemble bent bristles project forwards and 

 downwards from both upper and lower jaws and cannot be used for 

 biting in the ordinary sense and seeing that we find the same ar- 

 rangement of the teeth (in part at least) in Coccosteus which is one of 

 the oldest Palaeozoic fossils it is unlikely that Cbaetostomus has ever 

 possessed any other arrangement apart from the fact that every step 

 towards a biting mouth is away from the Myxinoids. This fact about 

 Coccosteus was first noticed by Hugh Miller *) and redescribed by 

 Traquair 2 ). 



There is another phenomenon of interest about these ancient 

 Placodermi. Asterolepis possesses a small maxillary bone so like that 

 of Siluroids, where it is situated at the base of a tentacle, that we 

 must draw the conclusion that these Palaeozoic forms were also pro- 

 vided with oral tentacles. 



Though not showing quite so primitive features as the Siluroids 

 the loaches, Cyprinoids, Motella 3 ) and other fish possess tentacles. 

 Misgurnus (Cobitis) fossilis possesses a large number (5 pairs), but the 

 only pair deserving special mention is the premaxillary pair which is 

 wanting in Siluroids. The nerve to this pair anastomoses with the 

 maxillary nerve but otherwise corresponds with the nerve of the pre- 

 maxillary tentacle of Myxine. It runs just below the eye stalk, the 

 eye in correlation with the presence of tentacles being small. 



The barbules of Sturgeons are premaxillary and maxillary tentacles 

 as shown by the nerve supply which I have followed in sections. 



Among the Anura the tadpoles of Dactylethra (Xenopus) possess 

 a maxillocoronoid tentacle and correspondingly a prepalatine piece is 

 present (upper labial [Parker]). I regard it as very probable that ten- 



1) Quoted in the biography of Louis Agassiz by E. C. Agassiz. 



2) Teaquair, Ann. and Mag. Nat. Hist. (6) 5, 1890. 



3) Zineone, Osservazioni anatomiche su di alcune appendici Aattili 

 dei pesci, in Rendiconti, Accad. Sc. Napoli, 1876. 



