380 



Fishery Bulletin 104(3) 



3 





■a 

 c 



t>D 



o * 



ca c 



y) bo 



bu t-^ 



3 C 



S c 



3 "> 



2 '■" 



O 



3 S 



CO CO 



O 00 0> to 00 



C^) ,M ,-H W ,-1 



CD 



oi io m 

 00 lr~ I~- 



CJJ CD CD CD CO 



g 2 s s 



,0-^-5-5 



CM LO CO 





cocococococooico 



-^ ^ c ;; 



ri. P -O ■« -S 





CO CO 



CO W CO CO CO CO t/3 



CO 

 CO 



3 

 C 

 OJ 

 be 

 J3 

 3 



bo 

 C 



c 



^ S 





CO 



~ o 3 



ll 



CO en 



marizes the meristic data used to identify 

 species morphologically. When a fish could 

 not be assigned to a species, it was assigned 

 to a species complex or subgenus. 



The individuals genetically identified to 

 species were six S. hopkinsi, four S. semi- 

 cinctus, two S. goodei, one S. auriculatus, one 

 S. jordani, one S. levis, one S. rastrelliger, 

 and one S. saxicola. The restriction fragment 

 patterns of these juveniles were identical to 

 those of previously observed adult specimens, 

 except for the two S. goodei juveniles, each of 

 which differed by a single difference (presum- 

 ably a site loss in one fish and a site gain in 

 the other) from reference specimens for Mbo 

 I (Table 3 and Li et al., 2006). The genetic 

 species assignment of all but one of these 

 specimens matched the identification based 

 on morphological characters (Table 4). The 

 discrepant fish was identified morphologically 

 as either S. seinicinctus or S. hopkinsi. The 

 specimen had 14 dorsal rays, which is within 

 the range of both species (Table 2) and was 

 a transforming pelagic juvenile that had not 

 yet developed the discernible body pigmenta- 

 tion pattern of older juveniles (Table 4). 



At the beginning of the study, S. oralis was 

 not included in our database. When we added 

 restriction site information for S. ovalis to the 

 database, we found that the species is very 

 similar to S. hopkinsi in its haplotype profile. 

 Consequently, three individuals morphologi- 

 cally identified as S. hopkinsi were genetically 

 assigned to the complex S. hopkinsi-S. ovalis. 

 The restriction enzyme Dde I can delineate the 

 two species; however, when we discovered the 

 ambiguity, no samples remained for analysis. 

 Based on meristics, the three specimens are 

 more likely S. hopkinsi than S. ovalis. The 

 range of counts for fin rays overlaps for the two 

 species (Table 2). However. S. ovalis common- 

 ly has eight or nine anal rays, and the three 

 specimens in question possessed seven rays 

 that occur commonly in S. hopkinsi (Table 4). 



Three specimens were assigned to the spe- 

 cies complex S. carnatus-S. chrysomelas-S. 

 caurinus according to morphological criteria. 

 Haplotype information eliminated S. cauri- 

 nus. Genetic analysis of the one specimen 

 that was morphologically assigned to the sub- 

 genus Sebastomus, which includes 13 North 

 Pacific species, reduced the possible species 

 to S. chlorosticus, S. eos, or S. rosenblatti. 



Discussion 



Our results demonstrated that restriction 

 site analysis of mtDNA is a simple and effec- 

 tive tool for identifying juveniles of Sebastes 



