BY J. DOUGLAS OGILBY. 751 



(3) According to Castelnau, " the head has no scales," but 

 those of the occiput must have been overlooked by him, for though 

 they are small and deeply embedded, they are nevertheless plainly 

 visible. 



These, however, are but minor discrepancies as compared with 

 (4) the dentition; referring to this Castelnau writes — "the teeth 

 extend on the vomer and the palatines ; the posterior 

 part of the tongue is also covered with them." This is quite the 

 opposite of what I find; in all my examples there is no sign of 

 teeth on any part of the mouth except those on the jaws. If 

 Castelnau's fish i-eally had the subsidiary teeth attributed to it by 

 its describer— which on a review of all the facts of the case I may 

 be permitted to doubt — it would of course be necessary to place 

 it in another genus; and this has possibly been already done b}^ 

 Dr. Bleeker, since his Pldli/pnodon nndiceps possesses the same 

 dentition as that assigned to his species by Castelnau.* 



The differences which sejoarate grandiceps from jitii/iceps are 

 undoubtedly slight, but those which are noticed in the preceding 

 analysis (see p. 746) appear to be constant ; the close affinity 

 between the two species was recognised by Castelnau, who writes: 

 " The principal reason for not uniting my sort with Krefft's is, 

 that he says that the pectorals attain the base of the anal; while 

 in my specimens they do not." I consider this elongation of the 

 pectoral fins to be merely a sexual character. 



This little fish is abundant in the Yarra, along the banks of 

 which it is known as the "Big-head"' according to Castelnau 



* The want of Bleeker's paper prevents me from ascertaining whether 

 his genus Philyjmodon is founded upon Castehiau's description of nudicepfi; 

 if tliis be the case, Bleeker's genus, being specially formed on account of a 

 character whicli it does not possess, must if monotypic be suppressed. And 

 this raises another question to which I am unable to find a satisfactory 

 answer, namely- -if a genus be founded on a character which is jjurely 

 mythical, should the name so proposed stand in preference to another 

 correctly characterised from the same species but at a later date? If the 

 practice of forming new genera from descriptions onlj- were discouraged or 

 disallowed, errors of this nature would soon cease. 



