LUVARID^E OZYMANDIAS 43 



part of the body are fairly shown. The vertebrae are strong, not longer 

 than deep, and without the peculiar structure called "trellised" in Auxis 

 and EUTHYNNUS. 



No XII is another torso, similar, but with the dorsal fin and the 

 preopercle obscured; depth about one and a half in head; no scales 

 shown; dorsal with about eight spines visible, the first opposite sixth 

 vertebra and opposite ventrals. A short or broken pectoral and ventral, 

 so distorted as to lie close together, the pectoral shorter than the snout, 

 the ventral a little shorter; lower jaw strong and projecting. 



No. X consists of two caudal fins relatively perfect, each with one 

 or two basal vertebras. The rays cannot be exactly counted. 



No. VIII is the posterior portion of a large fish, supposed to be 

 from the diatomaceous shale at Lompoc, loaned from the collection of 

 Miss J. M. Telford. Twelve vertebrae are shown and about ten dorsal 

 rays. The scales and other fins are obscure, but the species is evidently 

 the same as the TUNITA of El Modena. 



The genus TUNITA differs from Auxis, AUXIDES and SCOMBER, the 

 latter its nearest relative, in the close approximation of the two dorsal 

 fins, the first short and high, of eight spines only, and in the rapid 

 tapering of the body backward. From Auxis and EUTHYNNUS it 

 differs in the simple form of the vertebrae. 



We propose for this genus the name of TUNITA, a local Spanish 

 term, for the smaller tunnies (EUTHYNNUS). 



Family LUVARID^ (?). 12 



38. Ozymandias gilberti Jordan, new genus and species. 

 (Plate XXII) 



Type a large fossil collected at San Pedro, in rocks now regarded 

 as of Miocene age. No. CXXXII. 



This consists of part of a robust vertebral column of a fish which 

 must have been nearly six feet in length in life. These vertebrae pre- 

 served are about twenty-four in number. The largest is two and a half 

 inches long and one and a half in depth, decreasing in size backward 

 but not much in length. Traces of strong neural spines. There were 

 probably about forty vertebrae in life. The vertebrae are very robust, 

 more or less hourglass-shaped. No traces of head or fins remain. While 

 in a general way this suggests a scombroid fish, there are no data to 

 indicate that it belongs to LUVARUS or any other recognized genus. 



To draw attention to this large fish, I attach to the type the name 

 of OZYMANDIAS GILBERTI, the specific name in honor of its discoverer, 

 Dr. James Z. Gilbert. The generic name refers to the heroic giant 



12 LUVARUS species (?). Jordan: "Fossil Fishes of California," 1907, p. 134. 



