1904,] OSTEOLOGY OF CLUPEOID FISHES. 467 



to lie over- the aiiterior half of the ramus ; it is formed mainly, 

 but not entirely, by the clentary. The angular is distinct. There 

 are no teeth. 



Hyojmlatine Series (text-fig. 128, p. 466). — The hyomandibular 

 articulates with the cranium by two heads : a small anterior 

 and a broad posterior one. It is long, and its axis slopes forward 

 and makes an angle of about 110 degrees with the symplectic, 

 which is disposed almost horizontally. The extremity of a 

 triangular process which rises upward and forward from the 

 external face of the hyomandibular lies ovei- the outer surface of 

 the postfrontal spine, and forms with it a kind of sliding joint. 

 The palatine is very short and broad, and has a single head for 

 articulating with the ethmoid region of the cranium. There are 

 no teeth. 



OjJercidar Series (text-figs. 127 and 128, pp. 465, 466).— The 

 opei'cular and subopercular bones are of average proportions. The 

 branchiostegaJ rays are six in number ; the first three are rather 

 slender and closely set, and situated some distance in advance 

 of the other three, which are greatly expanded and overlap the 

 former thi-ee. The first four are attached to the outer edge of the 

 ceratohyal, the fifth lies over the suture between the ceratohyal 

 and epihyal, while the last is attached to the outer edge of the 

 epihyal. 



Ilyohranchial Series (text-fig. 129, p. 468). — The hyoid is short, 

 so short, indeed, that the end of the epihyal is only slightly behind 

 the end of the first hypobranchial. The lower hypohyal is about 

 twice as large as the upper. The glossohyal consists of a stout 

 rod of cartilage, with a small membrane-bone on the posterior 

 part of its upper surface. The urohyal is a strong bone, the 

 sections of the posterior half of which have the form of an 

 inverted Y, The third basibranchial is nearly as large as the 

 second, and the cartilage that represents the fourth and fifth 

 basibranchials is drawn out to a considerable length. 



Concerning Chatoessus Giinther states ('Study of Fishes,' 1880 

 p. 657) that the branchial arches form two angles, one pointing 

 forwards and the other backwards. This statement, which 

 appears to be based vipon the description given by Valenciennes 

 (Hist. Nat. Poiss. xxi. 1848, p. 96), does not, I believe, refer to 

 the skeletal arches as one might suppose on reading the passage, 

 but is an allusion to the remarkable continuation of the series of 

 gill-filaments from the top of the epibranchials, particidarly the 

 first, backwai'd along the side of the parasphenoid. 



The anterior ends of the fourth ceratobranchials are expanded 

 both forwards and backwards, so that while not relinquishing 

 their connection with the anterior ends of the fifth ceratobranchials, 

 they nearly touch the mesial ends of the third ceratobranchials. 

 The fifth ceratobranchials are considerably expanded, as also are the 

 fourth epibranchials. The latter have a curved wall of cartilage 

 extending upward and inward from the outer edge of the bone, 

 in relation with the epibranchial organ. 



30* 



