Kusaka et al.: Genetic population structure of Dipturus chinensis around Japan 
107 
Table 4 
Results from mismatch distribution analysis and neutrality tests for 3 clades, based on phylogenetic analysis of mito- 
chondrial cytochrome 6b gene (931 base pairs), of the polkadot skate (Dipturus chinensis). The demographic parameters 
estimated for each clade in the mismatch distribution analysis include time since expansion in units of mutational time 
(t) and population size before (89) and after (6,) a rapid change in population size in units of mutational time (Rogers and 
Harpending, 1992; Schneider and Excoffier, 1999). The goodness of fit between the observed and expected distributions was 
tested by using the sum of squared deviations (SSD) and the Harpending’s raggedness index (Hri). The neutrality of the 
sequence variation was verified by using Tajima’s D statistic and Fu’s Fg test. P-values are provided in parentheses. DNA 
sequences were obtained from specimens collected around Japan, in the East China Sea, Sea of Japan, and Pacific Ocean, 
during 2010-2017 or were obtained through the International Nucleotide Sequence Database Collaboration for populations 
in Taiwan or off the Korean Peninsula. 
Mismatch distribution 
Mismatch 
Group Tt observed mean 0) 0, 
Clade A 3.695 2.314 0.000 3.916 
Clade B1 3.000 0.427 0.000 0.299 
Clade B2 1.533 1.320 0.072 6.597 
of the Tsushima Current, passes eastward through the 
Tsugaru Strait and runs into the Oyashio Current east of 
the Tsugaru Strait, forcing it to flow southward along the 
eastern coast of northern Honshu. The distribution of the 
Tsushima lineage of various marine animals on the Pacific 
coast of northern Honshu is considered to be influenced by 
the Tsugaru Warm Current (Kojima et al., 2004; Akihito 
et al., 2008; Katafuchi et al., 2011; Kokita and Nohara, 
2011; Hirase et al., 2012). In contrast, a genetic differ- 
entiation between the western and eastern populations 
of northern Honshu has been suggested for the ocellate 
spot skate, the Pacific sandlance (Ammodytes personatus), 
and a zoarcid fish, Davidijordania poecilimon (Han et al., 
2012; Kai et al., 2014; Misawa et al., 2019b), as well as for 
the polkadot skate as indicated by our results. 
The dispersal of species is affected by many factors, 
including the environmental requirements and life histo- 
ries of species. Kai et al. (2014) suggested that gene flow 
between the Sea of Japan and the Pacific populations of 
D. poecilimon is restricted by the shallow Tsugaru Strait 
(which has a maximum depth of approximately 150 m) 
because this deep-sea species is found in depths of less 
than 262 m (Shinohara et al., 1996). The normal habi- 
tat depths of polkadot skate (mainly 20-150 m; Hatooka 
et al., 2013) and ocellate spot skate (mainly 30-100 m; 
Last et al., 2016) are shallower than that of D. poecili- 
mon, and their migrations are not as restricted by the 
shallow strait as that of D. poecilimon. The clear genetic 
divergence in polkadot skate and the restricted gene flow 
in ocellate spot skate (Misawa et al., 2019b) between the 
western and eastern populations of northern Honshu may 
be partly because of the lack of pelagic eggs and the larval 
stage of rajids (Ishiyama, 1967). 
For polkadot skate, results of our study indicate a 
divergence between the southern (clade B1) and northern 
(clade B2) populations on the Pacific coast of Honshu. The 
Test of goodness of fit 
SSD (P) 
0.007 (0.686) 
0.007 (0.334) 
0.006 (0.578) 
Neutrality tests 
Hri (P) Tajima’s D (P) Fu’s F, (P) 
—10.959 (0.002) 
—0.413 (0.248) 
—2.025 (0.089) 
0.022 (0.893) 
0.393 (0.572) 
0.047 (0.798) 
—1.518 (0.037) 
—1.679 (0.024) 
—0.858 (0.221) 
existence of geologically discrete lineages on this coast 
of Honshu has been suggested for some coastal fish spe- 
cies (Katafuchi et al., 2011; Han et al., 2012; Hirase et al., 
2012; Hirase and Ikeda, 2014). Among them, lineages of 
surfperch and goby species have been reported to be sepa- 
rated by the Izu Peninsula (Katafuchi et al., 2011; Hirase 
et al., 2012). Hirase et al. (2012) suggested that the shal- 
low coastal areas in bays east and west of the Izu Penin- 
sula, suitable for a goby species, likely disappeared owing 
to the sea level fall in the Pleistocene glacial periods, a 
change that resulted in vicariant separation. 
In contrast, Han et al. (2012) detected a boundary 
between the lineages in the transition zone between the 
Kuroshio and Oyashio Currents and suggested that sea 
temperature is a key factor preventing the dispersal of the 
lineage that dominates the northern group of Pacific sand- 
lance. The Kuroshio Current is subtropical and saline, 
and the Oyashio Current is subarctic and less saline 
(GSI, 1990). Results of the studies of the genetic popula- 
tion structure of mottled skate (Im et al., 2017; Misawa 
et al., 2019a) indicate that temperature gradients poten- 
tially affect the allopatric distributions of lineages. In 
thorny skate, genetic differentiation is reportedly caused 
by differences in salinity and temperature (Chevolot et al., 
2007). The divergence between the southern (clade B1) 
and northern (clade B2) populations of polkadot skate on 
the Pacific coast of Honshu was probably due to environ- 
mental differences, such as differences in temperature 
and salinity between the Kuroshio and Oyashio Currents. 
Evolutionary history 
For polkadot skate, the divergence time between clade A 
and clade B was estimated to be 1.31—2.10 MYA in the 
Early Pleistocene. Some changes in land area with respect 
to glacio-eustatic sea-level fluctuation occurred in the 
