TETRAGONURUS ATLANTICUS. 131 



tliyologists. Tlie close alliance naturally of Tclragomirus with tlie 

 Scombridal genera Thyrsites, or Aplurus (Escolar), may be inferred 

 both from its being called vernacularly Escolar de nature, and from the 

 true Scombridal Escolar having, though erroneously, been supposed to 

 belong to the same genus. 



Indeed there seems little either of external character or inward struc- 

 ture to warrant the arrangement of this fish amongst the Mugilida;. 

 These have few C(Eca, Tetragonurus having many : in the true Grey 

 Mullets the ascending branch of the stomach dilates into a sort of giz- 

 zard ; in Tetragonurus it is simple : they have few vertebrae ; Tetra- 

 gonurus has many : they have minute and feeble or setaceous teeth ; it 

 has them distinct and bony: they have the palatines and vomer normally 

 unarmed ; it has them armed : they have smooth and apparently entire 

 scales*; it has them rough and pcctinato-ciliate. In Tetragonurus the 

 form of the fins, especially of the two dorsal fins, and the nature of their 

 rays differ remarkably from the usual character of these parts in the Mu- 

 gilidce ; and the anatomical structure of the pharynx, mouth, and teeth, 

 has nothing whatever of the peculiar organization which characterizes so 

 singularly the Grey Mullets. The pseudo-abdominal or backward posi- 

 tion of the ventral a little behind the root of the pectoral fins, is a very 

 trifling, Mlacious character j- when resting as here, and in many other 

 Acanthoptcrygious fishes,^ on mere external appearances : and if insuf- 

 ficient for a ground of association with the Grey Mullets, it is absolutely 

 of no value whatever as a mark to separate Tetragonurus from the Scom- 

 bridee, Cuv. ; for the ventral fins, or their rudiments, in a crowd of 

 these, § are equally behind the pectoral, and are vastly more so in Nota- 



* A. high power of the microscope shows, indeed, that the edges of their laminae in Mugil are 

 finely creniilate : in which, if not sui generis, Xhey may be perhaps considered to approach rather to 

 the ctenoidal, than to be of the true cycloidal structure. But if this serve not for a caution against 

 employing too absolutely the intimate formation of the scales in founding or defining the main ich- 

 thine divisions, it shows at least the liability to practical uncertainty attendant on these characters. 

 In the Scombridce, nonnally the scales are truly cycloidal : in the Mugilidee, they are only so ap- 

 parently, or ambiguously : in Tetragonurus, otherwise Scombridal, they are eminently and distinctly 

 ctenoid. Thus the character in which Tetragonurus approaches, according to this theory, nearest to 

 the MugUidcB, is the very point in which it most appears, except on microscopical investigation, to 

 differ from them. And even after appeal to the microscope, it might truly be affirmed that there is 

 no stronger discrepancy between the scale of a true Scombridal fish and of a Muyil, than between 

 one of the latter and of a Tetragonurus. This genus may, however, be considered as in several re- 

 spects connecting MugilidcB with Scombridal, 



t See Cuv. and Val. Hist. iii. G7 ; il. xi. 195. — The ventral fins in Tetragonurus are not truly 

 abdominal or unconnected with the bones of the shoulder, as in the Abdominal Malacopterygians, 

 or the Acanthopterygian Sphjrasnidoe. 



% The ventral fins are quite as abdominal, apparently, in the typically thoracic genus Perca, as in 

 Tetragonurus and the Grey Mullets. 



§ E. g. Iiampris, Scomber, Tlnjrsiles, Genipiilaa ]>ruper, Lepidopus, Aplurus, some species of Liclda, 

 Caranoe, &e. 



