49 
Abstract— Species-specific restriction 
site variation in the 12S/16S rRNA 
and ND-3/ND-4 mtDNA regions was 
used to distinguish among 15 rockfish 
species of the genus Sebastes common 
to the waters of Alaska. Intraspecific 
variation exhibited by eight of the spe- 
cies (based on five individuals of each 
species) did not obscure the interspe- 
cific variation, except possibly between 
S. zacentrus and S. var-iegatus. Intra- 
specific nucleotide diversity averaged 
0.0024 substitutions per nucleotide, 
whereas interspecific nucleotide diver- 
gence averaged 0.0249. In contrast, the 
average nucleotide divergences between 
Sebastes and two other scorpaenid spe- 
cies, Helicolenus hilgendorfi and Sebas- 
tolobus alascanus, were 0.0805 and 
0.1073, respectively. Cladistic and phe- 
netic analyses supported some, but not 
all, of the subgenera assignments of 
Sebastes. A scheme for distinguishing 
among the species studied was pre- 
sented. Restriction sites of 10 restric- 
tion endonucleases were mapped in 
the two PCR-amplified mtDNA regions 
by using double digests. In all, we 
detected 153 sites corresponding to 640 
(13.5%) of the 4815 nucleotides in the 
two regions combined. The ND-3/ND-4 
region exhibited substantially more 
intraspecific, interspecific, and interge- 
neric variation than the 12S/16S rRNA 
region. 
Manuscript accepted 22 August 2000. 
Fish. Bull. 99:49-62 (2001). 
Identification of rockfish ( Sebastes sppJ 
by restriction site analysis of the mitochondrial 
ND-3/ND-4 and 12S/16S rRNA gene regions 
Anthony J. Gharrett 
Andrew K. Gray 
Fisheries Division, School of Fisheries and Ocean Sciences 
University of Alaska Fairbanks 
11120 Glacier Highway 
Juneau, Alaska 99801 
E-mail address (for A. J. Gharrett): ffaig@uaf.edu 
Jonathan Heifetz 
Auke Bay Laboratory 
Alaska Fisheries Science Center 
National Marine Fisheries Service, NOAA 
11305 Glacier Highway 
Juneau, Alaska 99801-8626 
The species-rich genus of Sebastes rock- 
fish has challenged both fisheries scien- 
tists and ichthyologists since they were 
first described from Alaskan waters by 
Tilesius (S. ciliatus\ 1813, cited in Esch- 
meyer, 1998) and Richardson (S. cauri- 
nus ; 1845). Both the large number of 
species, about 100 worldwide (Ishida, 
1984; Kendall, 1991), and the metamor- 
phoses that occur in larval and juve- 
nile fish produce a confusing number 
of forms. The diversity of species and 
forms combine to limit our knowledge 
of the biology, including life histories, of 
rockfishes. To date, identification to spe- 
cies is not possible for many larvae and 
juveniles (e.g. Kendall, 1991; Moser, 
1996), and distinguishing between some 
adult species may be difficult. For exam- 
ple, adult S. variegatus is similar to 
S. zacentrus and adult S. mystinus, S. 
melanops, and S. ciliatus are often mis- 
identified (Love 1 ). The inability to iden- 
tify species constrains surveys of larval 
abundance and, consequently, ecologi- 
cal studies that are important for con- 
servation and management of rockfish 
and other species. In addition, the ques- 
tions facing biologists and fishery man- 
agers require tools that can resolve 
intraspecific population (stock) struc- 
ture, as well as methods for identifying 
species. 
The size of the genus and the pau- 
city of information about some of the 
species have also contributed to a cha- 
otic history of their systematics and 
many aspects of the phylogeny have 
not been resolved (Kendall 2 and re- 
viewed in Cramer 11895] and Phillips 
11957]. Cuvier (1829, cited in Eschmey- 
er,1998) first described the genus Se- 
bastes for northern Atlantic specimens. 
The number of genera recognized for 
the species presently placed in Sebastes 
has expanded and contracted repeated- 
ly, reaching a maximum of 15 (Jordan 
et al., 1930) and now these genera are 
generally considered subgenera. When 
combined with five northwestern Pa- 
cific Ocean (Matsubara, 1943) and one 
northern Atlantic Ocean subgenus, Se- 
bastes comprises about 22 subgenera 
(Kendall, 1991). 
Identification and systematics of fish 
depend largely on morphological char- 
acters; morphology alone, however, does 
not always provide sufficient criteria, 
especially for identification of larval 
and juvenile forms. Genetic informa- 
tion, obtained by using biochemical or 
molecular methods, has been used to 
address systematic problems. In some 
1 Love, M. 2000. Personal commun. Marine 
Sciences Institute, University of California, 
Santa Barbara, CA. 93106. 
2 Kendall, A.W. 2000. Personal commun. 
NMFS Alaska Fisheries Center, 7600 Sand- 
point Way NE, Seattle, WA 98115. 
