Gharrett et al.: Identification of Sebastes spp. by restriction site analysis 
51 
Table 1 
Rockfish and related species and subgenera of Sebastes spp. used in mitochondrial DNA haplotype comparisons. The number des- 
ignates the species and the letter indicates the particular composite haplotype observed. 
Designation 
Common name 
Species 
Subgenus 
1, a and b 
Pacific ocean perch 
Sebastes alutus 
Acutomentum 
2, a and b 
rosethorn rockfish 
Sebastes helvomaculatus 
Sebastomus 
3 
quillback rockfish 
Sebastes maliger 
Pteropodus 
4, a and b 
redbanded rockfish 
Sebastes babcocki 
Rosicola 
5, a and b 
black rockfish 
Sebastes melanops 
Sebastosomus 
6 
yellowtail rockfish 
Sebastes flavidus 
Sebastosomus 
7, a-d 
sharpchin rockfish 
Sebastes zacentrus 
Allosebastes 
8 
harlequin rockfish 
Sebastes variegatus 
Allosebastes 
9 
redstripe rockfish 
Sebastes pronger 
Allosebastes 
10, a and b 
rougheye rockfish 
Sebastes aleutianus 
Zalopyr 
11, a and b 
yelloweye rockfish 
Sebastes ruberrimus 
Sebastopyr 
12 
shortraker rockfish 
Sebastes borealis 
Zalopyr 
13 
light dusky rockfish 
Sebastes ciliatus 
Sebastosomus 
14 
silvergray rockfish 
Sebastes brevispinis 
Acutomentum 
15, a and b 
copper rockfish 
Sebastes caurinus 
Pteropodus 
16 
helicolenus 
Helicolenus hilgendorfi 
17, a-d 
shortspine thornyhead 
Sebastolobus alascanus 
Table 2 
Primers used for polymerase chain reaction amplification of rockfish ( Sebastes , Helicolenus, and Sebastolobus spp.) mtDNA 
regions, a = Thomas and Beckenbach (1989); b = Cronin et al. (1993); c = Gharrett 7 ;d = Anderson et al. (1981); e = Anderson et al. 
(1982); f = Roe et al. (1985); g = Chang et al., 1994; h = Zardoya et al. (1995). 
Region amplified 
Sequence 
Location in O. mykiss h 
Source 
ND3/ND4 
5' TAACGCGTATAAGT G ACTT CCAA 3' 
5' TTTT GGTT CCTAAGACCAAT GG AT 3' 
bp 10574-10596 
bp 12881-12904 
from a (similar to b) 
from a and c (similar to b) 
12S/16S 
5' AATT CAGCAGT G ATAAACATT 3' 
5' AG ATAG AAACT G ACCT GG ATT 3' 
bp 1234-1254 
bp 3615-3635 
consensus: d, e, f, g 
consensus: d, e, f, g 
1 Gharrett, A. J. 2000. Unpubl. Oncorhynchus kisutch sequences. Fisheries Division, Univ. Alaska, Fairbanks, 11120 Glacier Hwy., Juneau, AK 
99801. 
Ultra Pure™ agarose [BRL Gibco, Grand Island, NY! 
and two parts Synergel™ [Diversified Biotech Inc., Bos- 
ton, MA]) in 0.5xTBE buffer (TBE is 90 mM tris-boric 
acid, and 2 mM EDTA, pH 7.5). DNA in the gel was 
stained with ethidium bromide and photographed on an 
ultraviolet light transilluminator. Digests that produced 
small unresolvable fragments on agarose gels were sub- 
jected to electrophoresis on 8 % polyacrylamide gels (29:1 
acrylamide:bisacrylamide) in 2xTAE (TAE is 40 mM tris- 
acetic acid and 1 mM EDTA, pH 8.0). DNA in poly- 
acrylamide was stained with SYBR Green 1 Nucleic Acid 
Stain™ (Molecular Probes, Eugene, OR). Molecular weight 
markers used to estimate restriction fragment sizes were 
100 base pair (bp) or 25-bp ladders (BRL Gibco, Grand 
Island, NY). Restriction sites were mapped by using dou- 
ble digests. Double digests were examined both in agarose 
and polyacrylamide by using 100- and 25-bp ladders. Com- 
posite haplotypes for all 10 restriction enzymes and both 
mtDNA regions were determined for each individual. 
Generalized (relaxed Dollo) parsimony trees (Swofford 
et al., 1996) were computed from shared restriction sites 
by a heuristic search with PAUP* 4.0 (Swofford, 1998), 
which assumed unordered states. Because the likelihood 
of the loss of a site is higher than the restoration of a 
lost site, we conducted analyses that assumed 1 ) no added 
cost, 2) twice the cost, and 3) four- times the cost for restor- 
ing a site. Multiple maximum parsimony trees from each 
analysis were combined to produce a majority consensus 
tree using PAUP* 4.0 (Swofford, 1998). A maximum-like- 
lihood tree was estimated with the program RESTML in 
