126 OSTAEIOPHTSI, 



Spiojysis of the Mexican and Central- American S])ecies. 



I. Middle of eye above tlie level of the angle of the mouth; 9 to 12 gill-rakers on the lower 

 part of the anterior arch. 



A. Aual fin of 19 to 2-4 rays. 



1. Occipital process not much broader than long. 



a. Teeth on the palate in two rather small, well-separated patches. 

 Diameter of eye 6 in the length of head (in specimens of 180 to 200 mm.) ; 



maxillary barbel extending to basal part of pectoral ; dorsal spine § the 

 length of head ; pectoral spine g the length of head, its inner edge very 



strongly serrated 1. melanopus. 



Diameter of eye 4 to 5 in the length of head (in specinaens of about 180 to 

 220 mm.) ; maxillary barbel extending about to base of pectoral; dorsal 

 spine I to J- the length of head : pectoral spine f the length of head . . 2. liropus. 



b. Teeth on the palate in two rather large narrowly separated 



patches ; diameter of eye 6 to 7^ in the length of head (in 

 specimens of 196 to 260 mm.) ; maxillary barbel extending to 

 basal part or middle of pectoral; dorsal and pectoral spines f 

 to I the length of head, the latter with the inner edge mode- 

 rately serrated 3. fuerthii. 



2. Occipital process much broader than long 4. evermanni. 



B. Aual fin of 27 rays 5. multiradiatus. 



II. Middle of eye about on a level with the angle of the mouth ; 30 gill- 

 rakers on the lower part of the anterior arch ; anal tiu of 22 or 

 23 rays 6. liypophthalmus. 



1. Arius melanopus. (Tab. XVIII. fig. 1 and Tab. XIX. fig. 7.) 



Arius melanopus, Glinth. Cat. Fish. v. p. 172 (1864) \ 



Tachysurus melanopus, Jord. & Everm. Bull. U.S. Nat. Mus, xlvii. 1900, p. 2784'. 



Galeichthys aqua-dulce. Meek, Publ. Columbian Mus., Zool. v. 1904, p. 9, t. 4 '. 



Depth of body 5 to 5i in the length, length of head 3| to 41. Breadth of head Ij in its length, diameter of 



on, viz., a single nasal opening on each side and a rayed adipose dorsal. Cuvier and Valenciennes (xv. p. 103, 

 1S40) placed Lacepudc's fish in the genus Pimeludus and made the following remarks : — " M. de Lacepede a 

 cru devoir faire un genre a part de ce poisson, a cause de quelques traits quo Ton voit sur I'adipeuse, ct qu'il a 

 regardes, dit-il, comme des rayons ; . . . . il le nommo tachisure a cause de I'agilite de sa queue, longue ct 

 doliiie, et comme on peut en juger par la copie, assez mauvaise d'ailleurs, qu'il donne de sa figure chinoise, les 

 proportions de cette partie ne different pas beauooup de oe qu'elle est dans les cspeoes du groupe aotuel. II 

 nous parait done que le genre Tachisure ne pcut etre conserve." 



I have thought it necessary to give in detail the reasons why I use Arius instead of Tacliysurus, as the 

 latter is now generally used by American ichthyologists, although I am at a loss to imagine how they have 

 determined that a fish known only from a Chinese painting had granular or obtusely conical teeth on the 

 palate. 



