Molecular and morphological differences between two geographic ...

2 downloads 0 Views 1MB Size Report
Choi, KC, Jeon, SR, Kim, IS, Son, YM 2002Coloured illustrations of the freshwater ... Kim, JK, Choi, JI, Chang, DS, Na, JT, Kim, YU 2003Distribution of fish eggs, ...
Molecular and morphological differences between two geographic populations of Salanx ariakensis (Salangidae) from Korea and Japan Jin Koo Kim1*, Ryu Doiuchi2, and Tetsuji Nakabo3 1

Fisheries Resources Research Team, National Fisheries Research and Development Institute, 408-1 Sirang-ri, Gijang-up, Gijang-gun, Busan, 619-902, Republic of Korea (e-mail: [email protected]) 2 Fisheries Experimental Station, Wakayama Prefectural Research Center of Agriculture, Forestry and Fisheries, 1551 Kushimoto-cho, Higashmuro-gun, Wakayama, 649-3503, Japan 3 The Kyoto University Museum, Kyoto University, Yoshida, Sakyo-ku, Kyoto 606-8501, Japan Received: January 6, 2005 / Revised: September 16, 2005 / Accepted: September 20, 2005

Ichthyological Research ©The Ichthyological Society of Japan 2006

Ichthyol Res (2006) 53: 52–62 DOI 10.1007/s10228-005-0315-1

Abstract Geographic variations of a salangid, Salanx ariakensis, from Korea and Japan, based on partial mtDNA cyt b gene sequences and morphometric data, are presented and compared with Salanx cuvieri from China. Analyses of molecular variance (AMOVA) suggested that the two populations of S. ariakensis differed significantly from each other in pairwise fixation index (FST = 0.43), although their reciprocal monophyly was not constructed in the phylogenetic analysis. The phylogenetic analysis, however, supported the reciprocal monophyly of S. ariakensis (Korean population + Japanese population) and S. cuvieri. Differences in genetic structure between the two populations of S. ariakensis were suggested by diversity indices, mismatch distribution shape, and minimum spanning network. In particular, diversity indices indicated that the Korean population was historically larger and more stable than the Japanese population. In addition, both the principal component analysis (PCA) and canonical discriminant analysis (CDA) for morphometric characters, and the Kruskal–Wallis test for some meristic characters, showed that the two populations of S. ariakensis differed from each other, although morphological differences between the two populations of S. ariakensis were smaller than those between S. ariakensis and S. cuvieri. Such morphological differences were consistent with the differences in mtDNA. Key words Salanx ariakensis · Cytochrome b · AMOVA · Morphological analysis

I

cefishes (noodlefishes) (family Salangidae), geographi cally restricted to East Asia where they occur in coastal seawater, rivers, and lakes, comprise 11 species (Roberts, 1984). Among them, Salanx ariakensis, which was described by Kishinouye (1901) from specimens from Ariake Bay, Kyushu, Japan, is known to have been widely distributed in northern to southern China, Vietnam (Zhang and Qiao, 1994), Taiwan (Oshima, 1919), along the western and southern coasts of the Korean Peninsula (Mori and Uchida, 1934; Wakiya and Takahasi, 1937; Kim, 1997; Choi et al., 2002; Kim et al., 2003), and in Ariake Bay (Hosoya, 2002). However, the population size of S. ariakensis is believed to have been reduced as a result of water pollution and/or overfishing along the coasts of China and Korea (Dou and Chen, 1994; Kim and Park, 2002). In addition, the species is regarded as critically endangered in Japan (Kawanabe et al., 2001). Accordingly, the biological characteristics, including geographic variation and aspects of the life history, should be investigated pending conservation efforts and stock management. Reports on the early life history of the species, i.e., Mizutani et al. (2000) and Hibino et al. (2002), were contributed for this purpose.

The present study aimed to identify genetic and morphological variations between geographic populations of S. ariakensis from Korea and Japan, based on partial mtDNA cyt b gene sequences and morphometric data. In this study, Chinese specimens of S. ariakensis were excluded because of unresolved taxonomic problems (Pan et al., 1991; Randall and Lim, 2000; Wang et al., 2001). Salanx cuvieri (known from China) was added with the geographic populations of S. ariakensis, because Roberts (1984) has indicated the close relationship of the two species. For the phylogenetic analysis, Salangichthys microdon was chosen as an outgroup.

Materials and Methods Samples.—The genetic analysis was based on 58 specimens (Korea, 28; Japan, 30) of S. ariakensis and 7 specimens of S. cuvieri with 1 specimen of Salangichthys microdon (outgroup). Sampling localities are shown in Fig. 1. For the morphological analysis, 49 specimens (Korea, 25; Japan, 24) of S. ariakensis and 24 of S. cuvieri were examined.

Geographic variations of Salanx ariakensis

Fig. 1. Sampling localities of Salanx ariakensis (solid circle) and Salanx cuvieri (solid rectangle)

Mitochondrial DNA sequence analyses.—Total genomic DNA was extracted from a piece of muscle tissue stored in 99% ethanol, using the DNeasy Kit (Qiagen, Hilden, Germany) and following the manufacturer’s protocol. The anterior half of the cyt b gene was amplified by polymerase chain reaction (PCR; Saiki et al., 1988), using the following primers: L-GLUDG (5¢-TGACTTGAARAACCAYCGT TG) + H-CB3 (5¢-GGCAAATAGGAARTATCATTC) (Palumbi et al., 1991). The PCR reaction fluid was a mixture of 8.3 ml sterile, distilled H2O, 1.2 ml dNTPs, 1.5 ml 10¥ buffer (Takara, Shiga, Japan), 1.5 ml each primer (5 mM), 0.1 ml Ex Taq, and 1.0 ml template DNA. PCR proceeded under the following conditions: initial denaturation at 94°C for 5 min, 30 cycles of denaturation at 94°C for 1 min, annealing at 54°–56°C for 1 min, and extension at 72°C for 1 min (final extension at 72°C for 5 min). PCR products, confirmed by 1% agarose gel electrophoresis, were purified by ExoSap-IT (Amersham-Pharmacia Biotech, Buckinghamshire, UK) and subsequently used for direct cycle sequencing using the BigDye Terminator Cycle Sequencing Kit (Applied Biosystems, Foster City, CA, USA). The primers used for sequencing were the same as those for the PCR. DNA sequences were edited and checked using DNASIS version 3.5 and aligned by CLUSTAL W version 1.7 (Thompson et al., 1994). Phylogenetic relationships were estimated by the neighbor-joining (NJ) algorithm with an option of Kimura’s (1980) two-parameter model (Saitou and Nei, 1987) using PAUP version 4.0b, according to the guidelines for choosing nucleotide substitution models by Kumar et al. (1993). Confidence in each node was assessed by 1000 bootstrap replicates. Haplotype and nucleotide diversities for each population of S. ariakensis and S. cuvieri were calculated using Arlequin 2.000 (Schneider et al., 2000). The minimum spanning network (MSN) on the basis of minimum sequence differences between haplotypes was calculated for each population of S. ariakensis. Subsequently, the minimum spanning tree (MST) was constructed

53

Fig. 2. Diagram showing body (A) and head (B) measurements. BH1c3, body height 1–3; BW1–2, body width 1–2; CPH, caudal peduncle height; CPL, caudal peduncle length; ED, eye diameter; HH1–2, head height 1–2; HL, head length; HW1–2, head width 1–2; IOW; interorbital width; ML, maxillary length; PAL, preanal length; PDL, predorsal length; PML, premaxillary length; POL, postorbital length; PPL, prepelvic length; SL, standard length; SnA, snout angle; SnL, snout length; SnW, snout width at 1/2 of snout; TL, total length; UJL, upper jaw length. Body width 1–2 (BW1–2) not shown. Definitions correspond to Materials and Methods

according to the manual for Arlequin 2.000, especially in the alternative connections among haplotypes. Pairwise fixation index (FST) between populations and species based on pairwise difference between haplotypes, and the historical demography of each population of S. ariakensis based on mismatch distributions, were also calculated using Arlequin 2.000, the significance of the FST values being tested by 100,000 random permutations. Morphological analyses.—Twenty-five body dimensions (Fig. 2) were measured by vernier caliper (to 0.01 mm) after fixation in 99% ethanol. Body widths 1 and 2 (BW 1 and 2; not given in Fig. 2) were measured at the anterior origin of the pelvic and dorsal fins, respectively. Fin rays, jaw, and palatine teeth were counted under a stereomicroscope. Vertebrae were counted from radiographs prepared by soft X-ray (Hitachi) in conditions of 20 kV, 3 mA for 60–120 s. Methods of measurements and counts followed Wakiya and Takahasi (1937), Roberts (1984), and Nakabo (2002), with some modifications. Differences in slopes and elevations of linear regressions of each morphometric character to standard length were investigated by multiple comparison of analysis of covariance (ANCOVA) with Tukey-like test (Zar, 1999) after log transformation. Each meristic character was statistically analyzed by the Kruskal–Wallis nonparametric test (Zar, 1999). Principal component analysis (PCA) and canonical discriminant analysis (CDA) were adopted to separate Salanx individuals, based on 25 morphometric characters after log transformation. All the statistical analyses were performed using SAS version 8.01. Specimens used in this study are listed below with “mr” for morphological analysis and “mt” for mitochondrial DNA analysis. Institutional abbreviations followed Leviton et al. (1985). Salanx ariakensis.—Korea: FAKU 88515–88524 [10 specimens, mr, mt, 70.3–91.9 mm standard length (SL), Jindo (Fig. 3A)], FAKU 88525 (1, mr, 84.7, Jindo), FAKU 88526–88539

54

J.K. Kim et al. Table 1. Frequencies of haplotypes among Korean Salanx ariakensis, Japanese S. ariakensis, and Salanx cuvieri Haplotype

Fig. 3. A Salanx ariakensis from Korea; FAKU88515, 89.5 mm SL. B S. ariakensis from Japan; FAKU88540, 93.4 mm SL

(14, mr, mt, 69.6–92.9, Jindo), FAKU 88902–88905 (4, mt, 76.1–86.5, Jindo); Japan: FAKU 88540–88563 [24, mr, mt, 64.0–102.2, Ariake Bay (Fig. 3B)], FAKU 88564–88569 (6, mt, 70.0–85.2, Ariake Bay). Salanx cuvieri (all from China).—FAKU 87794–87797 (4 specimens, mr, mt, 123.0–133.8 mm SL, Hong Kong), FAKU 87798 (1, mr, 125.0, Hong Kong), FAKU 87800– 87802 (3, mr, mt, 115.0–124.1, Hong Kong), FAKU 87803– 87804 (2, mr, 115.5–122.0, Hong Kong), FAKU 87806–87807 (2, mr, 110.1–124.1, Hong Kong), FAKU 87809–87813 (5, mr, 112.3–123.8, Hong Kong), FAKU 87816 (1, mr, 109.7, Hong Kong), FAKU 87818–87822 (5, mr, 118.6–125.9, Hong Kong). Salangichthys microdon.—FAKU 88570 (1 specimen, mt, 78.9 mm SL, Lake Mikata, Japan).

Results Molecular analyses. A 786-bp portion of the mtDNA cyt b gene was obtained from S. ariakensis, S. cuvieri, and Salangichthys microdon. The nucleotide sequence data reported here appear in the DDBJ/EMBL/GenBank nucleotide sequence databases with the accession numbers AB196848–196913. Among 28 specimens of Korean S. ariakensis, 23 haplotypes (K1–23) were defined, one being shared by 4 specimens (K1) and two by 2 specimens (K2 and K3) (Table 1). Among 30 specimens of Japanese S. ariakensis, 12 haplotypes (J1–12) were defined, one being shared by 10 specimens (J1), one by 7 specimens (J2), and three by 2 specimens (J3–5) (Table 1). Seven haplotypes were defined among 7 specimens of S. cuvieri (C1–7, Table 1). Haplotypes were shared neither between populations of S. ariakensis nor between S. ariakensis and S. cuvieri. The estimated ratios of transitions to transversions were 4.1 (33/8), 23 (23/1), and ND (15/0) in the Korean S. ariakensis, Japanese S. ariakensis, and S. cuvieri, respectively. The majority of substitutions occurred at the third codon position, except in five Korean specimens, in which they occurred at the first codon transition (C ´ T). Pairwise uncorrected genetic distances were 0%–1.78% (mean, 0.82%) and 0%– 1.53% (mean, 0.59%) in the Korean S. ariakensis and Japanese S. ariakensis, respectively, whereas those between

Number of individuals

Salanx ariakensis Korea (Jindo) K1 K2 K3 K4–23 Japan (Ariake Bay) J1 J2 J3 J4 J5 J6–12 Salanx cuvieri C1–7

4 2 2 1 10 7 2 2 2 1 1

Table 2. Mean genetic distances within and between Korean and Japanese populations in Salanx ariakensis and Salanx cuvieri Species (population)

S. ariakensis (Korea)

S. ariakensis (Japan)

S. ariakensis (Korea) S. ariakensis (Japan) S. cuvieri

0.009 ± 0.002 0.013 ± 0.003 0.081 ± 0.009

0.006 ± 0.002 0.085 ± 0.010

S. cuvieri

0.007 ± 0.002

Genetic distance values ± standard error are estimated by Kimura’s (1980) two-parameter model

populations were 0.25%–1.91% (1.25%) (Table 2). Pairwise uncorrected genetic distances within S. cuvieri were 0.38%–1.02% (0.62%). Mean genetic distance between the Korean + Japanese S. ariakensis and S. cuvieri was 7.81% (see Table 2). The Korean and Japanese S. ariakensis were almost entirely separated by a NJ tree based on the mtDNA cyt b gene, but some specimens from the two populations combined to form small clades (Fig. 4). On the other hand, genetic differences between the Korean population and the Japanese population in S. ariakensis were suggested by a high pairwise fixation index (FST = 0.43), being significant in the permutation test. Haplotype diversity (h) for the Japanese population was 0.841, being significantly less than values for the Korean population (0.979) and S. cuvieri (1.000) (t test, P < 0.05) (Table 3). Nucleotide diversity (p) was highest in the Korean population (8.2%) and lowest in the Japanese population (5.9%), S. cuvieri being intermediate (6.2%) (Table 3). The reciprocal monophyly of S. ariakensis and S. cuvieri was supported by a 100% bootstrap value (Fig. 4). In addition, they also showed a significant FST value (0.88).

Geographic variations of Salanx ariakensis

55

Fig. 4. Neighbor-joining dendrogram derived from Kimura’s (1980) distance matrix, based on the first 786 bp of mtDNA cyt b gene sequences among two populations of Salanx ariakensis and Salanx cuvieri, with Salangichthys microdon as an outgroup. K1–29, J1–30, and C1–7 indicate Korean Salanx ariakensis, Japanese S. ariakensis, and Salanx cuvieri, respectively. Numbers at branches indicate bootstrap probabilities (>50%) in 1000 bootstrap replications. Bar equals 0.001 of Kimura’s (1980) distance

According to the minimum spanning tree, the Japanese population in S. ariakensis included two prevalent haplotypes with a few connected by short branches and a few linked to the Korean population in S. ariakensis. However, the haplotypes of the Korean population were distributed widely (Fig. 5). According to the mismatch distribution shape for each population, that from Korea increased rapidly at a low substitution rate, peaked at 3 substitutions, and thereafter

slowly decreased with two subsequent peaks at 7 and 12 substitutions, whereas the Japanese population showed an early peak, but then decreased abruptly to zero; subsequently, a second peak occurred near 9 substitutions, resulting in a profile greatly different from that for the Korean population (Fig. 6). Morphological analyses. Meristic and morphometric characters of S. ariakensis and S. cuvieri are shown in Table 4. Multiple comparison results (ANCOVA) showed that the

56

J.K. Kim et al.

Table 3. Measures (±SD) of genetic diversity within and between two populations of Salanx ariakensis and Salanx cuvieri Species

Number of samples

Number of haplotypes

Diversity index Haplotype

S. ariakensis Korea Japan Korea/Japan S. cuvieri S. ariakensis/cuvieri

Nucleotide (%)

Mean pairwise differences

28 30

23 12

0.979 ± 0.018 0.841 ± 0.049

0.82 ± 0.45 0.59 ± 0.33

6.46 ± 3.15 4.66 ± 2.35

7

7

1.000 ± 0.076

0.0062 ± 0.0039

4.857 ± 2.694

FST

P

0.434

Observed SSD] = 0.147). However, estimated initial population sizes (q0) and mutational timescales (t) suggest that the Korean popu-

lation was historically a larger and older population than the Japanese population. Morphological differences were also found in both morphometric and meristic characters among the two geographic populations of S. ariakensis and S. cuvieri. Both the PCA and the CDA for morphometric characters showed that although the population of S. ariakensis from Korea differed slightly from that from Japan, it was considerably closer to the latter than to S. cuvieri. A similar conclusion was reached for meristic characters, including dorsal fin ray, anal fin ray, vertebra, presymphyseal, and dentary teeth numbers. These morphological dif-

Geographic variations of Salanx ariakensis

59

Table 5. Results of multiple comparisons of analysis of covariance (ANCOVA) for slope and elevation of morphometric characters in standard length with covariates between Korean Salanx ariakensis and Japanese S. ariakensis, and between S. ariakensis and Salanx cuvieri Species Area

S. ariakensis Korea(K)

Number of specimens 25 Standard length (mm) 70.2–92.8 Slope in SL Head length 1.07 Head height (HH1) 0.68 Head height (HH2) 0.85 Head width (HW1) 0.88 Head width (HW2) 0.89 Eye diameter 0.52 Interorbital width 0.86 Upper jaw length 1.17 Snout length 1.21 Snout width 1.05 Body height (BH1) 1.08 Body height (BH2) 0.81 Body height (BH3) 0.90 Body width (BW1) 1.20 Body width (BW2) 0.65 Predorsal length 1.07 Prepelvic length 1.06 Preanal length 1.06 Caudal peduncle 0.83 sheight Snout angle -0.01 Postorbital length 1.20 Premaxillary length 1.51 Maxillary length 1.07

Slope

Elevation

Comparisons

S. ariakensis

S. cuvieri

Japan(J)

Combined

China

24 63.9–102.2

49 63.9–102.2

24 109.7–133.8

1.15 1.18 1.11 1.02 1.09 0.72 1.14 1.25 1.24 0.96 1.20 1.07 1.19 1.59 1.51 0.99 1.02 1.02 1.14

0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0

0 2 0 0 1 2 0 0 2 0 2 2 2 2 — 1 1 0 2

0.02 1.22 1.52 1.12

0 0 0 0

2 2 1 0

J>K

J>K J>K

K>J J>K J>K J>K K>J J>K K>J K>J J>K J>K J>K K>J

Slopes

Elevation

Comparisons

1.11 0.93 1.03 0.96 0.99 0.59 1.09 1.25 1.28 0.95 1.10 0.89 1.03 1.53 1.26 1.02 1.04 1.03 0.94

1.16 1.09 0.90 0.81 0.86 1.59 0.67 0.73 1.02 0.89 0.61 0.65 0.74 0.77 0.69 0.97 0.94 1.03 0.99

0 0 0 0 0 1 0 1 0 0 0 0 0 1 0 0 0 0 0

2 2 1 0 2 — 1 — 2 2 0 0 2 — 0 0 0 0 0

A>C A>C A>C

-0.08 1.16 1.57 1.08

0.02 0.99 0.83 1.47

0 0 1 0

2 2 — 2

A> A> A> A>

A>C C>A A>C A>C C>A A>C

A>C A>C

C C C C

Numbers in slope and elevation indicate significance [0 (P > 0.05), 1 (P < 0.05), and 2 (P < 0.01)] BH1-3, BW1-2, HH1-2, and HW1-2 correspond to Fig. 2

ferences agreed closely with genetic differences already discussed. During the Quaternary period, the Tsushima Strait opened and closed in keeping with glacial–interglacial cycles, which might have influenced biogeographical patterns, owing to the formation and destruction of environmental barriers (Kitamura, 2004). The habitats of the Korean population and the Japanese population in S. ariakensis were likely to have been variously separated and connected each other at that time. It is estimated that the Korean population and the Japanese population shared a common ancestor within 2,889,000 years using an estimated rate of evolution of 0.45% per million years for the mtDNA cyt b gene (Minamoto and Shimizu, 2002). Therefore, the molecular and morphological differences between the populations may have resulted from the last geological event, since which time the Tsushima Strait has been open. Notes. Salanx ariakensis was initially described from specimens collected from Ariake Bay, Japan, being characterized by 13 dorsal fin rays, 24 anal fin rays, and 9 pectoral fin rays (Kishinouye, 1901; Kishinouye in Jordan and

Snyder, 1902). The present specimens agreed with the original description (Kishinouye, 1901), except for anal fin rays, which numbered 26–28. This difference may have occurred because Kishinouye did not count the anteriormost 2 minute anal fin rays. The genetic study suggested the Japanese population of S. ariakensis to be the most endangered. Especially, diversity indices and the initial population size of Japanese S. ariakensis were considerably smaller than those of Korean S. ariakensis. Allowing for its life span, estimated as 1 year (Roberts, 1984), its extinction may be complete. Acknowledgments We are grateful to D.S. Kim (Wando Maritime and Fisheries Office, Korea), J.Y. Kim (Jindo, Korea), K.W. Suzuki, S. Islam, M. Tanaka (Kyoto University, Japan), and D.-H. Chong (Hong Kong, China) for assistance in the collection of specimens in Jindo, Ariake Bay, and Hong Kong. We also thank Y. Kai (Kyoto University, Japan), M.S. Hwang (National Fisheries Research and Development Institute, Korea) and, G.S. Hardy (New Zealand), who reviewed the manuscript. This study was supported by the Post-doctoral Fellowship Program of Korea Science and Engineering Foundation (KOSEF).

(n = 24) (n = 24) (n = 24)

(13) (24) (21)

(20) (24) (22)

S. ariakensis Korea Japan S. cuvieri

S. ariakensis Korea Japan S. cuvieri

S. ariakensis Korea Japan S. cuvieri

0 2 1

6

0 1 0

1

4 3 0

1 2 1

7

11

0 3 0

2

4 1 2

8

14 19 4

12

4 2 0

3

7 5 6

9

6 2 16

13

5 7 3

4

0 0 4

14

(22) (24) (22)

10

3 4 7

3 2 7

6

0 2 5

7

0 0 0

7

2 3 3

11

1 4 0

12

Maxillary teeth

1 5 2

5

Palatine teeth

Dorsal fin rays

1 1 2

13

0 0 1

8

0 1 2

8

1 0 0

14

0 0 2

9

3 10 18

9

0 2 0

15

0 1 0

10

12 11 0

10

0 0 0

0 0 0

12

(21) (24) (23)

(21) (24) (22)

16

7 3 0

11

Pectoral fin rays

0 0 0

3

0 0 0

2 0 0

5

13

9 2 0

4

2 0 0

6

(16) (23) (22) 1 9 9

26

2 8 5

27

2 0 0

7

5 11 0

5

4 2 0

8

5 10 2

6

5 4 1

9

2 1 3

7

3 6 3

10

Premaxillary teeth

0 0 0

25

0 0 7

8

2 4 3

11

8 6 3

28

0 0 7

9

0 0 3

10

1 2 4

12

4 0 3

29

Anal fin rays

0 3 1

0 0 0

31

17 19 0

0

(21) (24) (22)

0 5 0

1

0 1 0

72

0 2 5

14

0 0 3

15

0 1 1

16

Dentary teeths

(17) (24) (17)

13

1 0 2

30

Table 6. Frequency distributions of meristic characters among Korean Salanx ariakensis and Japanese S. ariakensis, and Salanx cuvieri

3 11 0

74

9 5 0

75

7 1 4

0 0 0

2

0 0 1

17

0 0 0

3

0 0 0

18

0 0 6

4

0 0 0

19

0 0 6

5

0 0 0

20

0 0 2

6

76

Presymphyseal teeths

1 6 0

73

Vertebrae

0 0 0

21

0 0 2

7

1 0 11

77

0 0 1

22

0 0 1

8

0 0 7

78

60 J.K. Kim et al.

Geographic variations of Salanx ariakensis

61

Table 7. Results of Kruskal–Wallis test for meristic characters between two populations of Salanx ariakensis, and between S. ariakensis and Salanx cuvieri S. ariakensis

Dorsal fin rays Pectoral fin rays Anal fin rays Vertebrae Palatine teeth Premaxillary teeth Maxillary teeth Dentary teeth

Table 9. Standardized canonical (Can) coefficients based on 25 morphometric characters among Korean Salanx ariakensis, Japanese S. ariakensis, and Salanx cuvieri

S. ariakensis

S. cuvieri

Measurements

Can 1

Can 2

Total length Standard length Head length Head height (HH1) Head height (H2) Head width (HW1) Head width (HW2) Eye diameter Interorbital width Upper jaw length Snout length Snout width Body height (HH1) Body height (HH2) Body height (HH3) Body width (HW1) Body width (HW2) Predorsal length Prepelvic length Preanal length Caudal peduncle height Snout angle Postorbital length Premaxillary length Maxillary length Eigenvalue Proportion Cumulative

-14.948 12.931 0.171 -0.357 0.327 1.523 -1.243 -0.003 -0.055 1.139 2.295 -0.758 -1.748 0.020 0.114 0.929 0.615 3.801 1.594 -2.903 1.485 -1.314 -1.264 2.160 -1.365 41.020 0.935 0.935

1.255 2.226 -0.854 0.012 -0.674 0.856 -0.178 0.494 -0.855 2.561 0.828 0.110 -0.442 2.772 -1.338 -3.244 0.435 -4.869 -6.716 8.001 2.460 0.200 0.553 -2.219 -0.690 2.861 0.065 1.000

Korea

Japan

Combined

China

12.0a 10.2a 28.1a 75.2a 4.2a 5.0a 9.7a 8.4a

12.0a 9.7b 26.9b 74.0b 4.6a 5.4a 10.1a 11.0b

12.0a 10.0a 27.4a 74.5a 4.5a 5.2a 9.9a 9.8a

13.0b 9.1b 27.7a 77.1b 6.3b 8.3b 9.6a 13.2b

Values and different superscript letters indicate means and significance (P < 0.05), respectively

Table 8. Eigenvectors for the first three principal components (PC) based on 25 morphometric characters among Korean Salanx ariakensis, Japanese S. ariakensis, and Salanx cuvieri Measurements

PC 1

PC 2

PC 3

Total length Standard length Head length Head height (HH1) Head height (HH2) Head width (HW1) Head width (HW2) Eye diameter Interorbital width Upper jaw length Snout length Snout width Body height (BH1) Body height (BH2) Body height (BH3) Body width 1 (BW1) Body width 2 (BW2) Predorsal length Prepelvic length Preanal length Caudal peduncle height Snout angle Postorbital length Premaxillary length Maxillary length Eigenvalue Difference Proportion Cumulative

0.2120 0.2118 0.2125 0.1518 0.1915 0.2043 0.2098 0.1752 0.1885 0.2090 0.2086 0.1812 0.2062 0.1997 0.2033 0.2097 0.2070 0.2111 0.2104 0.2115 0.2051 -0.1653 0.1979 0.2077 0.1934 21.7816 20.6942 0.8713 0.8713

-0.1102 -0.1167 -0.0307 0.5389 0.2177 0.0991 0.0913 0.1036 0.1872 -0.1701 -0.1731 0.3213 0.0429 0.0756 0.1383 -0.0822 -0.0771 -0.1229 -0.1467 -0.1130 0.0352 0.5183 0.1189 -0.1874 0.0384 1.0873 0.6774 0.0435 0.9148

-0.0117 -0.0109 -0.0096 -0.2114 -0.3020 -0.0808 0.0398 0.7956 -0.1403 -0.0071 -0.0401 -0.0744 0.0669 0.0905 0.1293 -0.0816 -0.1128 -0.0167 -0.0098 -0.0004 0.1103 0.1762 -0.1674 -0.0455 0.2707 0.4099 0.0522 0.0164 0.9312

BH1-3, BW1-2, HH1-2, and HW1-2 correspond to Fig. 2

BH1-3, BW1-2, HH1-2, and HW1-2 correspond to Fig. 2

Literature Cited Avise JC (2000) Phylogeography: the history and formation of species. Harvard University Press, London, pp 135–211 Choi KC, Jeon SR, Kim IS, Son YM (2002) Coloured illustrations of the freshwater fishes of Korea (in Korean). Hyang-mun, Seoul, pp 147– 148 Dawson MN, Louie KD, Barlow M, Jacobs DK, Swift CC (2002) Comparative phylogeography of sympatric sister species, Clevelandia ios and Eucyclogobius newberryi (Teleostei, Gobiidae), across the California Transition Zone. Mol Ecol 11:1065–1075 Dou S, Chen D (1994) Taxonomy, biology and abundance of icefishes, or noodlefishes (Salangidae) in the Yellow River estuary of the Bohai Sea, China. J Fish Biol 45:737–748 Grant WS, Bowen BW (1998) Shallow population histories in deep evolutionary lineages of marine fishes: insights from sardines and anchovies and lessons for conservation. J Hered 89:415–426 Hibino M, Kinoshita I, Ohta T, Tanaka M (2002) Morphology of Ariake icefish Salanx ariakensis larvae in the Chikugo estuary. Jpn J Ichthyol 49:103–108 Hosaya H (2002) Salangidae. In: Nakabo T (ed) Fishes of Japan with pictorial keys to the species, English edn. Tokai University Press, Tokyo, p 298 Jordan DS, Snyder JO (1902) A review of the salmonoid fishes of Japan. Proc U S Natl Mus 24:567–593

62 Kawanabe H, Mizuno N, Hosoya K (eds) (2001) Freshwater fishes of Japan (in Japanese). Yama-kei, Tokyo, pp 84–85 Kim IS (1997) Illustrated encyclopedia of fauna and flora of Korea (in Korean), vol 37. Freshwater fishes. Ministry of Education, Seoul Kim IS, Park JY (2002) Freshwater fishes of Korea (in Korean). KyoHak, Seoul Kim JK, Choi JI, Chang DS, Na JT, Kim YU (2003) Distribution of fish eggs, larvae and juveniles around the Youngsan River estuary. J Kor Fish Soc 36:486–494 Kimura M (1980) A simple method for estimating evolutionary rate of base substitution through comparative studies of nucleotide sequences. J Mol Evol 16:116–120 Kishinouye K (1901) New species of Salanx found in Ariake Bay. Zool Mag (Tokyo) 13:359–360 Kitamura A (2004) Quaternary climatic changes and species diversity: responses of marine organisms to glacial-interglacial cycles. Fossils 75:54–59 Kumar S, Tamura K, Nei M (1993) Guidelines for choosing distance measures. In: MEGA: molecular evolutionary genetics analysis, version 1.01. Pennsylvania State University, University Park, pp 27– 29 Leviton AE, Gibbs RH Jr, Heal E, Dawson CE (1985) Standards in herpetology and ichthyology: Part I. Standard symbolic codes for institutional resource collections in herpetology and ichthyology. Copeia 1985:802–832 McCartney MA, Acevedo J, Heredia C, Rico C, Quenoville B, Berminghams S, McMillan WO (2003) Genetic mosaic in a marine species flock. Mol Ecol 12:2963–2973 Minamoto T, Shimizu I (2002) Phylogenetic relationships among osmerid and salangid fish inferred for mitochondrial cytochrome b gene sequences. Mem Fac Sci Kyoto Univ (Ser Biol) 18:1–13 Mizutani H, Matsui S, Takeshita N (2000) Embryonic, larval and juvenile development of Salanx ariakensis (Pisces: Salangidae). Suisanzoshoku 48:497–502 Mori T, Uchida K (1934) A revised catalogue of the fishes of Korea. J Soc Chosen Nat Hist 19:1–16 Nakabo T (2002) Introduction to ichthyology. In: Nakabo T (ed) Fishes of Japan with pictorial keys to the species, English edn. Tokai University Press, Tokyo, pp xxi–xlii

J.K. Kim et al. Oshima M (1919) Contributions to the study of the fresh water fishes of the island of Formosa. Ann Carnegie Mus 12:169–328 Palumbi S, Martin A, Romano S, McMillan WO, Stice L, Grabowski G (1991) The simple fool’s guide to PCR. University of Hawaii, Honolulu Pan JH, Zheng CY, Wu HL, Liu JH (1991) The freshwater fishes of Guangdong Province. Guangdong Science & Technology Press, Guangdong Quinn TW (1992) The genetic legacy of mother goose: phylogeographic patterns of lesser snow goose Chen caerulescens caerulescens maternal lineages. Mol Ecol 1:105–117 Randall J, Lim KKP (2000) A checklist of the fishes of the South China Sea. Raffles Bull Zool Suppl 8:569–667 Roberts TR (1984) Skeletal anatomy and classification of the neotenic Asian salmoniform superfamily Salangoidea (icefishes or noodlefishes). Proc Calif Acad Sci 43:179–220 Rogers AR, Harpending H (1992) Population growth makes waves in the distribution of pairwise genetic differences. Mol Biol Evol 9:552– 569 Saiki RK, Gelfand DH, Stoffel S, Scharf SJ, Higuchi R, Horn GT, Mullis KB, Erlich HA (1988) Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 239:487–491 Saitou N, Nei M (1987) The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol Biol Evol 4:406–425 Schneider S, Roessli D, Excofier L (2000) Arlequin version 2.000. A software for population genetics data. University of Geneva, Geneva Thompson JD, Higgins DG, Gibson TG (1994) Clustal W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res 22:4673–4680 Wakiya Y, Takahasi N (1937) Study on fishes of the family Salangidae. J Coll Agric Tokyo Univ 14:265–296 Wang S, Wang Z, Li G, Cao Y (2001) The fauna of Hebei, China. Hebei Science and Technology, Hebei Zar JH (1999) Biostatistical analysis, 4th edn. Prentice Hall, New Jersey Zhang YL, Qiao XG (1994) Study on phylogeny and zoogeography of fishes of the family Salangidae. Acta Zool Taiwanica 5:95–113