430 
Fishery Bulletin 109(4) 
southward to central Honshu on the Pacific coast of 
Japan and to the western coast of Honshu. Amaoka 
(1984) and Nakabo (2002b) recognized the three species 
as valid, following Chen and Barsukov (1976). Several 
authors recognized only S. vulpes — S. ijimae and S. 
zonatus being considered as synonyms (Kanayama and 
Kitagawa, 1983; Nagasawa, 2000; Kitagawa et al., 
2008; Ishida, 2009). However, because comprehensive 
genetic or morphological studies have been lacking to 
date, taxonomic status of the three species remains 
uncertain. 
Failure to recognize reproductively isolated popula- 
tions within an exploited stock can introduce critical 
errors in management (Carvalho and Hauser, 1994). 
Sebastes vulpes and S. zonatus are both abundant 
across northern Japan, together representing an im- 
portant fisheries component, whereas S. ijimae is rela- 
tively rare (Sekigawa et ah, 2003). Because of their 
high commercial value, some Japanese fisheries orga- 
nizations have attempted to enhance the stocks of S. 
vulpes and S. zonatus through aquaculture (Sasaki, 
2003; Sekigawa et al., 2003). However, without reliable 
taxonomic information for both species, it is unlikely 
that enhancement of the fishery will be realized. In 
this context, as a first step toward fully resolving the 
taxonomic status of S. vulpes , S. ijimae, and S. zonatus, 
we focused on S. vulpes and S. zonatus, using compre- 
hensive genetic and morphological analyses. 
Mitochondrial DNA (mtDNA) has been used success- 
fully as a primary marker to infer species boundar- 
ies among species of Sebastes (e.g., Alesandrini and 
Bernardi, 1999; Kai et al., 2002b), although recently 
evolved sibling species pairs that are nonmonophyletic 
with respect to the mtDNA gene tree have been fre- 
quently reported in Sebastes (e.g., Kai et al., 2002a; 
Narum et al., 2004; Burford and Bernardi, 2008). 
The delimitation of such pairs requires data from 
multiple independent loci (e.g., Nichols, 2001; Avise, 
2004). In fact, S. vulpes and S. zonatus are primar- 
ily distinguished by body coloration, which has been 
demonstrated as a good indicator of recent speciation 
in Sebastes (e.g., Kai et al., 2002a; Narum et al., 
2004; Hyde et al., 2008; Orr and Hawkins, 2008). 
Accordingly, a technique called AFLP (amplified frag- 
ment length polymorphisms) (Vos et al., 1995) has 
also been used, because it is a multilocus approach 
that produces hundreds of highly replicable indepen- 
dent dominant markers (Bensch and Akesson, 2005) 
and therefore estimates genetic divergence across the 
whole genome. Such an approach has successfully 
resolved the species boundaries and phylogenetic re- 
lationships among recently diverged species complexes 
in various organisms, in which mtDNA sequencing 
alone was less informative (e.g., Seehausen et al., 
2003; Mendelson and Shaw, 2005). Moreover, evaluat- 
ing a mtDNA gene tree against the background of a 
multilocus approach allows further discussion on the 
evolutionary relationships and histories among closely 
related species (e.g., Kai et al., 2002a; Hyde et al., 
2008; Burford, 2009). 
Figure 1 
Collection sites around the Noto Peninsula, Island of 
Honshu, Japan, for the 65 Sebastes spp. specimens exam- 
ined for genetic and morphological differences. Numbers 
below symbols represent sample sizes. 
Materials and methods 
Samples 
In order to clearly demonstrate intrinsic reproductive 
isolation between S. vulpes and S. zonatus, a total of 65 
specimens were collected from single locality around 
Noto, Ishikawa Prefecture, Japan (Fig. 1), thereby elimi- 
nating any geographical variations. The body coloration 
of each specimen was recorded with a photograph taken 
while the fish was alive or soon after death. Thirty-nine 
specimens with a grayish body were identified as S. 
vulpes, and the remaining 26 (with a brownish body) as 
S. zonatus (Fig. 2), generally by following the methods 
of Chen and Barsukov (1976). The two species are not 
usually caught together; S. vulpes is caught with set 
nets at ~60 m depth and S. zonatus with gill nets at 
-150 m around the sampling locality (K. Sakai 1 ). Muscle 
tissue was taken from each specimen before fixation and 
preserved in 99.5% ethanol. The specimens examined 
here were deposited in the Kyoto University Fish Col- 
lection (FAKU) (see Appendix for catalog numbers and 
collection data). 
Genetic analysis 
Genomic DNA was extracted from the preserved muscle 
tissue, by using the DNeasy Tissue Kit (Qiagen, Tokyo, 
Japan) according to the manufacturer’s protocols. 
AFLP profiles were generated with the AFLP Plant 
Mapping Kit (Applied Biosystems, Foster City, CA) by 
following the manufacturer’s protocol slightly modified 
by Kai et al. (2002a). For the selective amplification 
1 Sakai, K. 2011. Personal commun. Noto Marine Center, 
3-47 Ossaka, Noto, Ishikawa 927-0552, Japan. 
