408 ATHEEINID^. 



not fixed to the humeral arch, which is never the case in the Trichi- 

 uridce with developed ventral fins ; the Trichiuridce, moreover, have no 

 true scales. I think, therefore, that Cuvier was right in placing this 

 fish near those Acanthopterygians which have abdominal ventral fins ; 

 but it is much more closely allied to Atherina than to Mugil, not 

 having eithier the mouth, or the pharyngeal apparatus, or the peculiar 

 structure of the stomach, or the ostcological peculiarities of the latter. 

 On the other hand, Tetragonurus agrees with Atherina in the struc- 

 ture of the gUls and of the fins, in the large number of vertebrae, 

 and in the natiu-e of its food, whilst the differences between the two 

 genera are such as to admit of their separation into two groups of the 

 same family. 



1. Tetragonurris cuvieri. 



Courpata (Nice). 



Mugil niger, Rondel, xv. cap. 6. p. 423; Gesner, Aquat lib. iv. p. 653; 



Willughby, p. 276, tab. R. 4. fig. 4. 

 Corvus niloticua, Akh-ov. De Pise. lib. v. c. 25. p. 610. 

 Tetragonurus cuvieri, Risso, Ichthyul. Nice, p. 347. pi. 10. fig. 37 (bad), 



and Mir. Merid. iii. p. 382 ; Cuv. 8f Val. xi. p. 172. pi. 318 ; Cuv. 



RhgneAnivi. III. Pom.pl. 76. fig. 2; Guichen. Explor. Alger. Poiss. 



p. 68; Mettenhcimer in Abhandl. Senck, Gesellsch. p. 215. tab. 17-19. 



B. 5. D. 15-21 1 ,-^. A. 10-12. L. lat. 118-120. 

 L. transv. 30. Vert. 36/22. 



a. Fine specimen. Sicily, 



b. Adult : stuffed, Mediterranean. Presented by "W. Yarrell, Esq. 



Mr. Lowe, who obtained the species in the sea of Madeira, points 

 out several differences between his specimen and the description of 

 Valenciennes, which appear to him of sufficient importance to di- 

 stinguish a second species, Tetragonurtis atlantmis (Proc. Zool. Soc. 

 1839, p. 79, and Trans. Zool. Soc. iii. p. 5, Pish. Mad. p. 130). Although 

 I do not share this opinion, it will always be valuable, in so scarce a 

 fish as T. cuvieri, to note peculiarities of the single individuals which 

 are preserved in our collections. 



The specimen which is preserved in spirits in the British Museum 

 is 12| inches long. The height of the body, behind the pectorals, 

 is contained seven times and a half in the total length ; the tail, 

 between the vertical fins, is rather deeper than thick ; the two keels 

 on each side of the caudal strongly prominent. The length of the 

 head is six times and a quarter in the total (the caudal fin included). 

 The eye is at least as high as broad, one-sixth of the length of the 

 head, and two-thirds of the width of the interorbital space. The 

 teeth in both jaws are curved backwards, triangular and compressed, 

 those of the upper jaw less than of the lower ; there are twenty- 

 eight on each intermaxiUary and thirty-three on each mandibulary. 



The spinous dorsal fin in both specimens is continued on to the 

 soft portion, and composed of twenty-one spines in the stuffed one 

 (as in Cuvier's figure), and of eighteen in the other, llisso and 

 Valenciennes found fifteen. Each of these spines is slender, as long 

 as tlie eye, with a membrane posteriorly, which docs not extend to 



