Misawa et al.: Population structure of Okamejei kenojei 
31 
35 n 
„ 30 
_i 
H 
£ 25 
o) 20 
C 
0 ) 
& 15 
a 
w 
« 10 • 
O 
>EC GYS ♦ SJ a EK 0 OS xNP 
a 71 
er 
x §*^xx 
xx x 
♦ x 
♦ x x 
♦ ♦ X 
X 
100 
300 
400 
500 
600 
200 
Total length (mm) 
Figure 5 
Plots of clasper length (as % of total length) of males from 
6 regional populations of Okamejei kenojei. EC=East China 
Sea; YS=Yellow Sea; SJ=Sea of Japan; EK=East coast of Ky¬ 
ushu Is.; OS=Osaka Bay; NP=Pacific coast of northern Japan. 
analysis; our results supported the uniqueness of 
the regional populations and suggested limits to 
migration. Such fine-scale population structures 
in other skate species have already been well 
documented (e.g., Chevolot et al., 2006; Griffiths 
et al., 2010, 2011; Dudgeon et al., 2012; Im et al., 
2017; Vargas-Caro et al., 2017) and are consid¬ 
ered to have been derived from the constraints 
of low dispersal because of large benthic egg cap¬ 
sules and the absence of a pelagic larval stage. 
Vargas-Caro et al. (2017) suggested that smaller 
skates inhabiting coastal areas had well-defined 
population structures than larger skates inhabit¬ 
ing offshore areas because body size and habitat 
are related to dispersal potential. Maximum size 
of members of the family Rajidae vary from 33 
cm to more than 2 m TL, and most members oc¬ 
cur from continental shelves to more than 4000 
m depth (Last et al., 2016). Studies on migration 
and habitat preference of O. kenojei are few; how¬ 
ever, the species is a typical small skate matur¬ 
ing at ca. 400-500 mm TL and inhabits coastal 
areas shallower than 150 m depth (Ishihara et 
al., 2009; Hatooka et al., 2013; Last et al., 2016). 
In comparison with the OS population, how¬ 
ever, morphological divergence of the NP popu¬ 
lation was somewhat remarkable against genetic 
differentiation; both genetic and morphological 
data sets did not agree completely. Among 12 
measurements of characters (excluding TL), as 
well as nuchal thorn counts, maturity size, and 
coloration, NP individuals were distinguished by 
8 of the 15 characters (disc width, dorsal head 
length, distance between orbits, distance between 
nostrils, distance between 1st gill openings, ma¬ 
turity size, nuchal thorn, and coloration), but 
OS individuals were distinguished by 6 of the 
15 characters (dorsal snout length, eye diameter, 
ventral head length, ventral snout length, prena¬ 
sal snout length, and nuchal thorn) (see Tables 
3-4; Figs. 3-7). It should be noted that we exam¬ 
ined a single maternally inherited gene (mtCR) 
which cannot be considered derived from male- 
mediated gene flow. This single gene may also 
explain the partial mismatch between our mtCR 
analysis and our morphological comparisons. Fur¬ 
thermore, genetic differences may also be ran¬ 
dom to some extent; Spies et al. (2006) indicated 
that genetic differences at mtDNA cytochrome c 
oxidase subunit I (COI) were not observed even 
among different species. In another case, incon¬ 
gruence between genetic and morphological varia¬ 
tion may be due to adaptations to different en¬ 
vironments. For example, adaptive evolutionary 
changes in life history, physiology, and phenotype 
in the winter skate (Leucoraja ocellata) have 
been associated with epigenetic regulation that 
causes changes in gene expression for adaptation 
to different environments without obvious genetic 
change (Kelly and Hanson, 2013; Lighten et al., 
