CRANIAL OSTEOLOGT OF TUK OSTEOGLOSSID.K, 263 



these, four are on the ceratohyal, one lies on the junction between 

 the epihyal and the ceratohyal, two on the lower edge o£ the 

 epihyal, and the last one about halfway up the outer face of tlie 

 epihyal. The first five are slender rods, the other three are 

 lamellate. The last of all, which is the largest, tends to remain 

 on the inner surface of the interojiercular when the head is 

 dissected prior to the maceration of the bones. Ruppeil (Beschr. 

 neu. Nil-Eische, 1829, p. 11) puts the number of branchiostegal 

 rays oi Heterotis as 7, as also do Hyrtl (Denkschr. Akad. Wiss. 

 Wien, viii. 18.54, p. 75) and Boulenger (Ann. Mag. Nat. Hist. 

 (7) viii. 1901, p. 51.5); Valenciennes {I.e. p. 471) and Giinther 

 (Brit. Mus. Cat. Fishes, vii. 1868, p. 380) say 8. Eight are 

 shown in pi. 8 (zootnm.), fig. 2 of Hemprich and Ehrenberg 

 (Symbolai Physicae, 1899). 



ELyohrancliial Series. — The hyobranchial skeleton of the 

 specimen examined is incomplete. The urohyal is short and stout 

 and lies between the strongly-developed processes that project 

 downward from the anterior ends of the second hypobranchials. 

 There is a small upper hypohyal, but no lower. The basibranchial 

 teeth, which are not quite so large as those of the entopterygoid, 

 are arranged in a patch, about as broad as long, set on the middle 

 part of the length of the membrane-bone that covers the hinder 

 part of the glossohyal, the whole of the first and second basi- 

 branchials and the anterior three-fourths of the third. The 

 epibranchial organ is borne by the fourth and fifth arches. 



The glossohyal, the first basibranchial, and the first pharyngo- 

 branchials appear to be cartilaginous, and there are no spicular 

 bones. According to Cope (Trans. Amer. Phil. Soc. n. s. xiv. 

 1871, p. 455) there are but two superior pharyngeals in Seterotis, 

 whereas in Osteoglossum there are three. Seeing that in both 

 Osteoglossum and Seterotis the first pharyngobranchial is un- 

 os^ified, the only explanation of the statement lies, I think, in 

 the fact that Cope, in examining Osteoglossum, mistook for a 

 pharyngobranchial the patch of coalesced teeth that lies behind 

 the third pharyngobranchial, and immediately beneath that car- 

 tilaginous rod which represents the fourth pharyngobranchial. 

 The dentigerous plate in question is wanting in Seterotis. 



Aeapaima gigas. 

 The only published figures and description of the skull of 

 Arapaima are those of Agassiz (Spix, Selecta Genera et Species 

 Pise. Bras. 1829, pi. B of the Anatomical Plates, and pp. .33-40). 



