246 MEMOIRS OF THE CARNEGIE MUSEUM. 



In our judgment, the Caristiidce compose a family distinct from the Ptera- 

 clidce, their nearest relatives, and from the Bramidce to which both are allied. The 

 apparent differences between Caristius and related genera may be thus indicated: 



Family PTERACLIDiE. 



Dorsal very high, beginning on head, continuous, its rays all. simple; anal similar, beginning far 

 forward, almost as long as dorsal; ventrals jugular, small, inserted below the eye. Scales firm, lobate. 

 Vertical fins with a basal sheath of scales. 



Pleraclis Gronow {velifera).^ 



Centropholis Hilgendorf (petersi). 



Dflrsal with 10 to 13 graduated rays; ventral rays, 1, 5; anal very high, its first long ray reaching 

 to end of base of fin. 



Plerycombus Fries (brama). 



As in Pleraclis, but the anal lower, the first long ray of anal reaching only to middle of fin. 



Benteuia Jordan and Snyder {(esticola) . 



Essentially as in Pteraclis, but with the fifth dorsal and second anal ray much enlarged. Ventrals 



very short, under the eye. 



Family CARISTIIDyE. 



Anal beginning near middle of body; ventrals jugular, but behind eye. Scales small, caducous. 

 Caristius Gill and Smith (japonicus). 



Family BRAMIDyE. 

 Characters included above. 



The Brainidw differ in the hard scales, and in the backward insertion of the dorsal which is low 



and behind the head. Anal moderate, inserted below middle of body. 



Family VELIFERIDiE. 

 The Veliferidm (Vilifer Schlegel), (hypselopierus), have the fins much as in Caristius, but the mouth 

 is wholly different, and the family has no relation to Caristius or to Pteraclis. 



Family PEMPHERID.E. 



119. Pempheris japonicus Doderlein. 

 Misaki, No. 6176a. 



Family CHEILODIPTERID^. 



120. Amia sialis, sp. nov. (Plate XXVIII, fig. 2). 



The following is a description of the type. No. 6021, C. M. Cat. of Fishes, 

 a specimen 123 mm. in total length, coming from Suruga Bay, Japan. 



Dorsal VII-I, 9; anal II, 8; scales in lateral line 25; in transverse series 10; 

 head 2.6 in body-length to base of caudal; depth 2.4; eye 3.75 in head; depth of 

 caudal peduncle 2.5 in head; gill-rakers 5+13. 



Body deep, nearly equal in depth to length of head; maxillary reaching to below 



2 Pteraclis ocellatus Cuvier and Valenciennes, with but two graduated rays in the dorsal and the ven- 

 trals I, 3, if these characters are exact, belongs to a different genus. It is probably, however, a true Pteraclis. 



J 



