792 PROCEEDINGS OF THE NATIONAL MUSEUM vol. xxx. 



Family GERRID.E. 



63. EUCINOSTOMUS CALIFORNIENSIS (Gill). 

 Two speciiiien.s from (Tiuiyaqiiil, Ecuador. 



64. GERRES PERUVIANUS Cuvier and Valenciennes. 



Several specimens were taken at (luayaquil, ditiering from speci- 

 mens from Panama only in l)eing darker in color. 



Family KYPHOSID^. 



65. DOYDIXODON L^EVIFRONS (Tschudi). 

 Plate LXVl, %. 2. 



A single specimen from ]Molendo, Peru, 27 cm. in length. 



This species may be known from D. fremininllei" by the produced 

 anterior rays of the soft dorsal forming an angle, which when depressed 

 reaches to the tip of the last dorsal ray. The fourth ra}' is the Icmgest 

 and forms the tip of the angle, behind which the posterior margin of 

 the fin is stronglj^ concave. 



In D. freminiriUei (specimens from the (lalapagos Islands in the 

 Stanford University collections) the soft dorsal is not angulated; the 

 tip of the fourth raj' is opposite the beginning of the last two-fifths or 

 one-third of the base of the fin. The fin is usually rounded and every- 

 where convex as shown in the accompanjnng figure, but its margin 

 ma}^ sometimes form a sigmoid curve, convex in front and concave 

 behind, and nowhere angulated except at tip of last ra}'. The latter 

 condition is shown in Valenciennes' plate,* and in the largest of our 

 specimens, -ta cm. in length, but the fourth ray is little if anj' longer 

 than when the fin is everywhere convex. This condition is prol)ably 

 developed with age. 



The anterior rays of the anal of D. Isevifrons are longer than in the 

 other species, making the posterior margin of the fin more obli([ue. 



Perhaps a greater difl'erence than these is shown in the size of the 

 teeth, which in D. Imvifrohs 2i\:% nearly twice as large as in 1>. fri- 

 rahivUlel, and are in fewer rows. In the former species the}' are in 5 

 oblique series, on the mandible, running downward and inward toward 

 the s3miphysis. In D. freminvUlei they are in 9 oblique series. 



The dorsal of our specimen of D. hriufrons has 15 rays. Of the 10 

 specimens of I). freminvUlei counted, 10 of th(>m have 17 ra3^s, 4 have 

 18 rays, and 2 have 10 rays. This is opposite to the condition alleged 

 to exist. Tschudi counts 18 rays in the type of I). Jievlf7'ons from 

 Huacho, Peru, and Valenciennes counts 15 rays in the t^'pe of D. 



"Plate LXVI, fig. 1. ''Voyage Venus, pi. v. 



