FISHES OF CEARA, BRAZIL JORDAN AND BRANNER 27 



Family OSTEOGLOSSID^ ( ?) 

 Genus CEARANA Jordan and Branner, new genus; type, Cearana rochae 



A specimen, badly preserved in a sand nodule, of different and 

 harder texture than most of the others from Ceara, seems to repre- 

 sent a new genus, which we refer very doubtfully to the Osteo- 

 glossidc-e, because, as in Ostcoglossuni, there is no division between 

 the opercle and subopercle. The elongate body distinguishes this 

 genus from Pharcodiis (Dapedoglossus) and Brych^tus; fossil 

 genera of the Eocene, referred to the Osteoglossidae. 



In Cearana the head is oblong, forming about two-sevenths of the 

 length to base of caudal. The greatest depth of the body is a little 

 less; the body is oblong; the distance from the gill opening to the 

 dorsal is considerably more than the greatest depth, a character ap- 

 parently important in this group. About two-fifths of the length of 

 the head is formed by the very large convex opercle, which is a single 

 undivided bone marked by radiating striae ; preopercle with the upper 

 limb erect and forming nearly a right angle. Jaws and teeth not 

 preserved, and mouth apparently large and oblique, extending past 

 the eye. Two postorbital bones behind eye and one on cheek approx- 

 imately subequal in size ; vertebrae 50. Distance of ventrals from 

 gill opening about equal to length of head. Dorsal short, rather 

 high, inserted over ventrals, its basal bones strong ; ventrals midway 

 between gill opening and anal. Caudal rays fine, the fin well forked, 

 the tail a little heterocercal. Scales not well preserved, apparently 

 small, firm and somewhat bony, with marked striae. 



II. CEARANA ROCH.Si Jordan and Branner, new species 



Plate VIII, Fig. 2 



Of this species we have two specimens (No. 5 and No. 32. Rocha 

 Collection), one in a nodule and represented in counterpart and one 

 small one not in a nodule. In one specimen (No. 5) part of the 

 body and the posterior portion of the head are very badly preserved. 

 The head must have been a little less in length than the distance from 

 the gill opening to the ventral fin. The preopercle is rounded, its 

 upright portion nearly vertical. The opercle is very large, very con- 

 vex, and in one piece, without separation of the subopercle. Its 

 length is about equal to its depth and about three-fifths the greatest 

 depth of the body, which is two-thirds the distance from gill opening 

 to central. Surface of opercle nearly smooth. Pectoral fin placed 

 low, a little longer than the opercle. Dorsal few-rayed (8 to 10) and 

 rather high, the first rays longest, the first interneural large and 



