INTRODUCTION. IX 



are silvery-lead. The head and opercles are plain ; not armed with spines 

 or teeth, as in Percida. 



Examples : — The Spet of Languedoc, Luzzo or Lucio di mare of 

 the Italians ; Barracuda of the West Indies ; Bicuda of Madeira {Esox 

 sphi/raua, L.). 



7Viglidt^: — The Gurnard tribe (Les Jones cuirassees, Cuv.) : latterl}^, by 

 Cuvier, separated from the Perches, with which they have most characters 

 in common, on account of the cuirass-like expansion of the bony plates 

 about their cheeks and head, giving them a peculiar physiognomy. 



Examples : — The Gurnards, Piper, River Bullheads, Sticklebacks of 

 England ; and the Cabra, Carneiro, Requeime, &c. of Madeira. 



Scicenidte: — The Maigre or Umbrina tribe. 



This group, as at present constituted, is rather a subordinate series 

 of PercidfE than a truly distinct family ; differing in no respect but in 

 the absence of palatal teeth, and offering also, in parallel order, precisely 

 the same combinations of other characters. Even the cavernous structure 

 of the head-bones is neither universal nor peculiaro It is wanting in a 

 number of genera ; whilst it is present in other fishes placed in other 

 families : e. g. in Trachichthys, placed amongst Percidte, and Monocentris 

 in Triglida;, by Cuvier himself. The absence of palatal teeth is even less 

 to be relied on ; as is proved by the same illustrious ichthyologist having 

 been compelled to place amongst Percidfe, not Trachicht/iT/s only, in 

 which they are feeble or evanescent, but several other genera, in which 

 they are as completely wanting as in any of his Sciantda. These teeth 

 are also wanting in Sj/nanceia, placed amongst Triglida, in which they 

 are generally present. 



But the great objection against keeping up this family in its present 

 form as hitherto defined, is, that it causes the separation of genera so 

 closely allied in other respects that they are scarcely distinguishable ; and 

 this, in deference only to a single artificial character, which is itself most 

 variable, and of very little value, either in itself, or as indicative of more 

 important differences of habit, food, or structure : a fact which might be 

 proved by instances innumerable. Thus Hamulon, Pristipoma, &c. are 

 separated far from their natural Percidous allies, Centropristis or Dules, 

 Therapo7i, Datnia, Pelates, and Helotes : whilst, as if to stamp more 

 strikingly the perfect futility of such a separation, founded on the presence 

 or absence of palatal teeth alone, the three latter genera of Percida have 

 absolutely none ; and in Therapon they are actually present or absent in 

 different individuals of the same species. 



Several genera of small fishes, (Amphtprion, Premnas, Pomacentrus, 

 Dascyllus, Glyphisodon, Etroploji, and Heliastes,) placed by Cuvier at 

 the end of his Scianidce, and characterized by their short oval shape, scaly 



