THE SHORE FISHES OF PERU 207 



nected by membrane, the third and fourth longest, scarcely as long as 

 snout and eye, reachmg origin of second dorsal if deflexed; second 

 dorsal and anal similar, each with an elevated lobe anteriorly, the 

 longest rays of dorsal about equal to postorbital part of head, each 

 with a detached finlet; origin of anal nearly an eye's diameter behind 

 that of second dorsal, the two fins coterminal; ventral fairly narrow, 

 inserted under or more usually slightly behind base of pectoral, 1.9 

 to 2.0 in head; pectoral long, falcate, reaching about opposite origin 

 of anal, 4.9 to 5.2 in length. 



Color bluish gray above; this color merging gradually into the pale 

 silvery of the lower parts; a dark spot on margin of opercle above its 

 posterior angle; ventral and anal pale; other fins more or less dusky; 

 axil of pectoral and inner surface of gill covers black. 



The foregoing description is based on sev^en rather large specimens, 

 555 to 565 mm. (485 to 497 mm. to base of caudal) long, which were 

 taken with hook and line at Callao and San Lorenzo Island. Some 

 small specimens, 73 to 150 mm. (60 to 117 mm. to base of caudal) long, 

 from Lobos de Afuera and Callao, collected by R. E. Coker, also are 

 on hand. The enumerations based on the small specimens come 

 within the scope of the larger ones. However, the proportions, espe- 

 cially as to the length of the head and depth of the body, differ so 

 greatly that it has seemed desirable to list them separately. Head in 

 length 3.3 to 3.5; depth 4.5 to 4.8; pectoral 3.85 to 4.3. Eye in head 

 3.35 to 3.6; snout 3.35 to 3.6; interorbital 4.3 to 4.6; maxillary 2.6 

 to 2.8; caudal peduncle 8.25 to 9.0; deepest scute in curved part of 

 lateral line 5.7 to 6.3, in straight part 4.7 to 6.3; ventral 1.85 to 2.0. 

 It is evident from the proportions that the body is deeper in the young, 

 in which it also is much more compressed. This character, then, is not 

 of much value in separating species, unless specimens of equal length 

 are compared. The scutes in the lateral line already are well devel- 

 oped in the smallest specimen, and the pectoral fin although propor- 

 tionately shorter aheady is long and somewhat falcate. The procum- 

 bent spine of the dorsal, which is more or less embedded in large 

 specimens, is free and sharp in the young, and the two anterior anal 

 spines already are free and detached from each other. 



Although I am unable to use Nichols's key (1920, p. 481) in separa- 

 ting this apparently valid species from T. symmetricus (Ayers), 

 from California, other characters do seem to distinguish it. First, and 

 apparently most important, the southern species, T. murphyi, has 

 a few more gUl rakers on the lower limb of the first arch, 45 to 48, 

 whereas the northern species, T. symmetricus, has only 40 to 42 

 (counted in six specimens from California). Second, the scutes in the 

 lateral line are deeper and more prominent in Peruvian fish, the 

 deepest in the curved and straight parts of the lateral line being of 

 about equal depth, whereas those in the curved part are narrower in 



