28 FOSSIL FISHES OF SOUTHERN CALIFORNIA 



Dorsal fin beginning about midway of length to base of caudal; its 

 height about equal to its base, the anal apparently smaller and weaker; 

 pectoral inserted low, in advance of posterior border of opercle, its rays 

 broken ; ventral fins mostly obliterated ; inserted about under middle of 

 dorsal ; caudal broken, apparently forked. 



This species apparently belongs to the genus ELLIMMA. From 

 ELLIMMA ELMODEN.E it seems to differ in the notably shorter head, and 

 caudal region, the larger body cavity, stronger ventral serratures and 

 fewer caudal vertebrae. The vertebras are 27+17 = 44 in ELLIMMA 

 ELMODEN^ and 16 -|-17 = 33 in ELLIMMA BRANNERI, the Brazilian 

 Eocene type of the genus ELLIMMA. 



24. Alisea grandis Jordan and Gilbert, new genus and species. 

 (Plate XVII, fig. 3) 



Type No. CXXXV, from San Pedro, in hard, flinty sandstone of 

 Miocene age. 



It is the posterior part of a large shadlike fish, the fragment pre- 

 served being 23 inches long, the whole fish in life upwards of 2^4 feet. 



Body elongate, compressed, no trace of protuberant ventral region ; 

 ventral scutes, if present, all obliterated. 



Insertion of dorsal fin nearly midway from nape to base of caudal, 

 the depth at that point 2% times in distance from insertion of dorsal to 

 base of caudal fin. As the anterior region is wanting the different meas- 

 urements may be compared with the greatest depth. Length of base of 

 dorsal uncertain, the posterior part obliterated. 



Anal fin better preserved, its insertion about under tenth ray of 

 dorsal, at distance from front of dorsal about two-fifths of depth of 

 body; the haemal spines, however, being very oblique, their insertion 

 much behind (five or six vertebrae behind) insertion of dorsal neurals; 

 length of anal fin from first ray to the last ray shown, about equal to 

 depth of body. Anal rays not far from thirty, the longest (anterior) 

 two and a half in depth of body. 



Vertebrae moderate, subequal, hourglass-shaped. Caudal vertebrae 

 twenty-eight, fourteen indicated in advance of these, the total number of 

 vertebrae not far from fifty. Last visible anal ray attached to the sixteenth 

 vertebra from the last, the fin covering about sixteen vertebrae ; caudal 

 vertebrae with strong, very oblique, neural and haemal spines. Hypural 

 obscured, apparently much as in a shad. 



Caudal fin very long and widely forked with numerous accessory 

 rays, the upper lobe pointed, its length one-fourth or more greater than 

 depth of body; lower lobe broken. 



Scales very small, smooth, cycloid, present on most of the body and 

 scattered in the rock about the back. The number was not far from 



