Cynoglossus joyneri Günther, 1878

Luo, Hai-Rong, Kong, Xiao-Yu & Munroe, Thomas A., 2023, Re-evaluation of the taxonomic status of four nominal, western Pacific species of tongue soles (Pleuronectoidei: Cynoglossidae: Cynoglossus), with redescription of C. joyneri Günther, 1878, Zootaxa 5360 (3), pp. 385-408 : 396-403

Redescription of Cynoglossus joyneri Günther, 1878

Figs. 1 , 4–5 ; Tables 5–6

Cynoglossus joyneri Günther 1878:486 ; Otaki 1896:421; Jordan & Starks 1907:241; Wu 1932:154; Fowler 1934:216; Chabanaud 1951:269; Chyung 1954:487; Zheng 1955:301; Zheng 1962:1011; Ochiai 1963:88; Zhang & Wang 1963:542; Cheng & Weng 1965:100; Menon 1977:46; Miao 1984:559; Shen 1984:140; Lu & Wu 1986:337; Shen et al. 1993:579; Kim & Choi 1994:805; Li & Wang 1995:361; Cheng 1997:463; Munroe 2000:646; Yamada 2000:1391; Ho & Shao 2011:62; Shen 2011:765; Kottelat 2013:466; Yamada & Yagishita 2013:1696; Fricke et al. 2017:84; Zhou et al. 2017:847.

Cynoglossus (Areliscus) lighti Norman 1925:270 ; Wu 1932:155; Chabanaud 1951:270; Kamohara 1953:9; Zheng 1955:300; Ochiai 1963:90; Cheng & Weng 1965:101; Liu & Qin 1987:441; Li & Wang 1995:363; Cheng 1997:464; Yamada 2000:1391; Yamada & Yagishita 2013:1696.

Areliscus tenuis Oshima 1927:201 .

Trulla lighti . Fowler 1934:216.

Cynoglossus tshusanensis Chabanaud 1951:270 .

Areliscus joyneri . Chyung 1954:487; Matsubara 1955:1286; Tanaka 1955:125.

Lectotype ( Figs. 1 , 4 ): BMNH 1878.4.15.94 ; Tokyo, Japan.

Diagnosis: Cynoglossus joyneri is a medium-sized tongue sole (≤ 333 mm SL) distinguished from congeners by the following combination of characters: caudal-fin rays 10; dorsal-fin rays 98–117; anal-fin rays 79–92; total vertebrae 49–54 (abdominal vertebrae 9; caudal vertebrae 40–45); middle lateral-line scales 62–80; scale rows between dorsolateral and midlateral lines 9–13; eyes small, diameter 4.1–11.5% of HL; head length (18.6–24.7% of SL) equal to or slightly longer than head width (17.6–24.9% of SL); two nostrils on each side of snout; three complete ocular-side lateral lines, no lateral line on blind side; ctenoid scales on both sides of body; ocular side of body and fins uniformly reddish-brown.

Description: Data for meristic and morphometric features are presented in Tables 5–6 . Dorsal-fin rays 98–117; anal-fin rays 79–92; pelvic-fin rays 4; caudal-fin rays 10; total vertebrae 49–54 (abdominal vertebrae 9, caudal vertebrae 40–45); middle lateral-line scales 62–80; scale rows between dorsolateral and midlateral lines 9–13.

Body moderately elongate, strongly compressed laterally; body depth (BD) 20.5–27.9% of SL, maximum body depth located at point between verticals through anus and body midpoint. Head relatively long (HL= 18.6–24.7% of SL) and wide (HW= 17.6–24.9% of SL). Head width, slightly subequal to slightly larger than, head length (HW/ HL= 0.9–1.2). Snout slightly rounded or bluntly pointed, snout length (SNL) 34.3–49.8% of HL; snout shorter than postorbital head length (POL= 43.5–65.3% of HL). Posterior tip of rostral hook not reaching, reaching, or extending slightly beyond, vertical through anterior margin of lower eye (PHE). Eyes subequal, anterior margin of upper eye slightly in advance of anterior margin of lower eye; eyes not contiguous, separated by narrow, nearly flat, interorbital space (IOW= 2.2–7.4% of HL), with 1–2 rows of minute ctenoid scales at its narrowest point. Mouth sub-terminal; length of upper jaw (UJL) 15.1–27.5% of HL; distance between angle of mouth to opercular margin (AMO) variable (39.1–60.2% of HL); ocular-side mouth cleft nearly straight, blind-side mouth cleft more semicircular. Angle of mouth located about at vertical through middle of head, extending posteriorly below or slightly beyond posterior margin of lower eye, sometimes lower jaw extending beyond posterior margin of lower eye by distance almost equal to one eye diameter. Lips on both sides smooth without fringed papillae. Teeth absent on ocular-side jaws; blind-side jaws with narrow band of small, villiform teeth. Upper head lobe (UHW= 10.0–20.2% of SL) wider than width of lower head lobe (LHW= 6.9–11.7% of SL). Preanal length (PAL) 20.2–28.0% of SL. Posterior margin of opercle with distinct indentation at, or near, its midpoint. Gill membranes united ventrally; free from isthmus. Gill arches without gillrakers.

Dorsal-, anal-, pelvic-, and caudal-fin rays soft, unbranched. Dorsal-fin origin near tip of snout. Anal-fin origin just posterior to vent. Unpaired (blind side) pelvic fin with membranous connection to first (anteriormost) anal-fin ray. Caudal fin pointed, its length (CFL) 20.5–27.9% of HL. Three lateral lines on ocular side; middle lateral line nearly straight along its length and ending at base of caudal fin or extending onto caudal-fin base by not more than 2 scales; dorsal and ventral lateral lines extending posteriorly along dorsal and ventral contours of body and usually ending at base of posteriormost 4–10 fin-rays of dorsal and anal fins, sometimes extending almost to distal ends of dorsal- and anal-fin rays. Dorsal and middle lateral lines connected by supraorbital commissure. Preorbital line on anterior snout present or absent. Cephalodorsal line well developed along anterior margin of snout and ending at posterior tip of rostral hook, connected to supraorbital line and preorbital line. Preopercular and mandibulo-opercular lines separated from each other; mandibulo-opercular line ending at or near posterior margin of opercle. No lateral line on blind side. Scales ctenoid on both sides of body, including those on lateral lines, except cycloid scales present anteriorly on blind side of snout. Genital papilla a short tube connected to first anal-fin ray.

Pigmentation: Background coloration of ocular side variable, either unevenly or uniformly brown, yellowish brown, reddish brown, or grayish brown. Fins uniformly pigmented with same color as, or with slightly lighter or darker color than, that on body. Oral cavity without conspicuous pigmentation. Outer surfaces of opercle on both sides of head with corresponding coloration as that on each side of body, respectively. Inner surfaces of both opercles dusky or dark brown. Skin behind gills on either side more or less darkly pigmented. Isthmus white. Peritoneum on both sides of body darkly pigmented, usually a dark brown. Blind side of body white.

Distribution: Cynoglossus joyneri is widely distributed in coastal marine waters of the northwestern Pacific Ocean, including the southern Sea of Japan, the Yellow and Bohai seas, East China Sea and South China Sea ( Günther 1878; Norman 1925; Wu 1932; Chabanaud 1951; Kamohara 1953; Ochiai 1963; Cheng & Weng 1965; Ochiai 1984; Kim & Choi 1994; Kim et al. 2005; Li & Wang 1995; Munroe 2000; Choi et al. 2002; Shen 2011; Yamada & Yagishita 2013; Munroe 2021).

In Japanese waters, C. joyneri occurs in coastal waters of the southern Sea of Japan and in Pacific coastal waters from Honshu to Kyushu. It is a common species reported from many localities, including Wakasa Bay, Suruga Bay, Osaka Bay, Seto Inland Sea, and off Kochi ( Günther 1878; Kamohara 1953; Ochiai 1963, 1984; Minami & Tanaka 1992; Baeck et al. 2011; Yamada 2000, 2002; Yamada & Yagishita 2013).

Off South Korea, this species has been reported from various locations in the southern Sea of Japan, the Yellow Sea, and in the East China Sea off southern Korea ( Lee 1989; Kim & Choi 1994; Hwang et al. 1998; Choi et al. 2002; Baeck & Huh 2004; Lee et al. 2010; Kwak et al. 2012; Yoon et al. 2013).

In Chinese waters ( Fig. 2 ), this species is known from a variety of locations ( Norman 1925; Wu 1932; Chabanaud 1951; Li & Wang 1995; Munroe 2021), including the Bohai Sea ( Zheng 1962; Shuozeng 1995a,b), coastal waters of the Yellow Sea ( Zheng 1955; Gu et al. 2009), the East China Sea, and the South China Sea ( Zheng 1962; Zhang & Wang 1963; Munroe 2000; Yamamoto et al. 2009), including the Yangtze River estuary ( Ruijing et al. 2010, Jiang et al. 2014; Song et al. 2015). It also occurs at Hong Kong ( Ni & Kwok 1999, and literature cited therein) and Hainan Island ( Lu & Wu 1986; this study).

Cynoglossus joyneri is also known from Taiwan ( Oshima 1927; Wu 1932; Chen & Weng 1965; Shen 1984; Shen et al. 1993; Shao et al. 2008; Shen 2011; Shao 2023). Reports of this species from off the south coast of Viet Nam ( Nguyen & Nguyen 2006) and Cambodia (based on museum specimens accessed through the FishNet2 portal) require further verification.

Cynoglossus joyneri lives on the continental shelf at an approximate depth range of 8–58 m ( Shuozeng 1995b; Jiang et al. 2014; citations in Munroe 2021).

Remarks. Taxonomic status of C. lighti in previous works and comparison with the features of C. joyneri . Previous studies ( Norman 1925; Wu 1932; Fowler 1934; Chabanaud 1951; Kamohara 1953; Zheng 1955; Ochiai 1963; Cheng & Weng 1965; Li & Wang 1995; Cheng 1997; Yamada 2000; Yamada & Yagishita 2013) considered C. lighti to be distinct from C. joyneri . An assortment of nine morphological characters ( Tables 1–4 ), including MLL, DMLL, BD/SL, HL/SL, HL/HW, PHE, size of lower eye relative to snout, snout shape (pointed or rounded), and presence/absence of a preorbital lateral line, were variously regarded in these studies as diagnostic for distinguishing C. lighti from C. joyneri . Our comparisons ( Table 5 ) revealed that values for 20/25 features examined, including eight of nine characters purported to be diagnostic for these two species in previous studies, varied only slightly and this amount of variation is indicative of intraspecific differences.

Features differing significantly between those in our study versus those in previously published works ( Table 5 compared with Tables 1–4 ) are the MLL counts recorded by Chabanaud (1951), and values of five other characters (MLL, CFR, BD/SL, HL/SL, and SNL/HL) reported in different studies ( Günther 1878; Fowler 1934; Chabanaud 1951; Zhang & Wang 1963; Menon 1977; Kim & Choi 1994). Possible explanations for variation between our results and those reported in historical literature arise from differences in how MLL scales were counted, especially with respect to beginning and end points for this count. Among previous records of MLL values for C. joyneri , 82–85 scales were reported for the two syntypes in Chabanaud’s (1951) work, whereas 85 scales (2 syntypes) and 78–86 scales (82 non-type specimens) were noted in the works of Günther (1878) and Kim & Choi (1994), respectively. In this study, and in most previous studies (which included a total of more than 321 specimens), MLL counts of 62–80 scales are reported for C. joyneri ( Table 5 ). Using the third method of counting MLL scales, we counted 80 MLL scales based on a photograph of the lectotype of C. joyneri . This count was similar to or included in the values reported for C. joyneri in the works of Chabanaud (1951), Günther (1878), and Kim & Choi (1994). Thus, the variation in reported counts of MLL for C. joyneri between our study and these others likely results from different counting methods for this feature, rather than from any actual differences.

Chabanaud (1951) recorded counts of 90–92 MLL scales for the lectotype and paralectotypes of C. lighti ( Table 3 ), which is distinctively different from the MLL counts (73, 84, and 79) we estimated using the three different methods for counting MLL scales described above. Chabanaud’s counts also differed significantly from the MLL values (75–78) reported for C. lighti in the original description ( Norman 1925; see Table 1 ). Given such disparity in the MLL counts between that of Chabanaud’s study (1951) and those reported in other studies, it is uncertain how Chabanaud counted MLL scales. Possibly, the counts he reported were in error.

Counts for CFR of C. joyneri have also varied in previous studies ( Tables 1 , 3 ), ranging from 8 to 11 fin rays ( Liu & Qin 1987; Li & Wang 1995). In these different studies, CFR counts have varied from 9–10 fin rays (based on 144 specimens), 8–10 (96 specimens), 10 (63 specimens), 8–11 (25 specimens), 10–11 (14 specimens), 8–9 (2 specimens) and 8 (1 specimen), respectively ( Tables 1 , 3 , 5 ). We counted 10 CFR for the lectotypes of both C. joyneri and C. lighti , and also for 133/138 non-type specimens examined in this study. Considering the symmetrical arrangement of caudal-fin rays in members of the Cynoglossidae and the relative conservative nature of this count, as well as the fact that 10 and 9–10 CFR were counted in our study and most previous studies, respectively, we consider 10 CFR to be the typical count for this feature in C. joyneri .

Values for some morphometric features have also varied between studies. For example, HL measurements expressed as a percentage of SL for C. joyneri were 18.6–24.7% based on data combined from more than 473 specimens included in our study and previous works where identifications were deemed reliable ( Tables 4–5 ). These values differ from those (HL= 18.9–40.0% of SL) recorded by Menon (1977) based on nine specimens, including the holotype of C. tshusanensis (see Table 2 ). Likewise, the range of values for BD/SL (20.5–27.9%) in the present study is much lower compared with that (20.3–43.6%) reported in Menon (1977). We can offer no explanation for such differences in head length and body depth measurements between those in our study and most previous literature compared with the values for these features listed in Menon (1977). Menon did not provide data for individual fish examined in his study, so it is impossible to identify the particular specimen(s) responsible for these higher values. Therefore, we consider the HL/SL and BD/SL measurements of 40.0% and 43.6% reported in Menon (1977) as extreme outliers or possibly as erroneous altogether.

Lastly, values of SNL/HL ranged from 34.3–49.8% of HL in our work ( Table 5 ). This range of values for C. joyneri is larger than that (27.8%) appearing in Lu & Wu (1986) based on a single specimen ( Table 2 ). We are unsure how this feature was measured by Lu & Wu, and thus, we consider their estimate of SNL/HL as an outlier value.

Several previous studies have examined molecular variation in specimens identified as C. joyneri and/or C. lighti . However, it is difficult to compare our results with those of these other studies ( Xu et al. 2008; Liu et al. 2010; Miao et al. 2013; Song et al. 2015; Wang et al. 2018). For example, Xu et al. (2008) and Liu et al. (2010) used different markers, such as 16S r RNA, ITS1 and Cytb. Liu et al. (2010) reported that C. joyneri and C. lighti could be possible synonyms but made no statement on the impact of this result regarding the taxonomic status of these nominal species. Song et al. (2015) also found only minor differences (0.007) between specimens they identified as C. joyneri and C. lighti , but made no statement on the impact of this result on taxonomic status. Also, these authors ( Song et al. 2015) did not submit sequences to GenBank so it is impossible to compare their sequences with those obtained in our study. Other studies, such as that by Wang et al. (2018), used the same COI marker as that in our study and concluded that they could not separate C. joyneri and C. lighti using this marker. Again, these authors did not elaborate on the taxonomic impact of this finding. Also, these authors did not provide morphological data for their specimens, so it is uncertain what criteria they used to identify their fish.

Values for all eight morphological characters purportedly diagnostic for distinguishing C. lighti from C. joyneri in previous studies demonstrated no interspecific differences between these nominal species when a large series of specimens were examined.Additionally, data from several molecular studies, including our study, and using a variety of different markers, also demonstrate that only a single species is represented among the material examined in these studies. Thus, the overall morphological and genetic similarities between both nominal species strongly support the hypothesis that these specimens are conspecific, and that C. lighti should be regarded as a junior subjective synonym of C. joyneri .

Taxonomic status of C. tenuis and C. tshusanensis . As mentioned above, previous works have regarded C. tenuis as a junior subjective synonym of either C. joyneri or C. lighti , and C. tshusanensis has been considered a junior subjective synonym of C. joyneri . After checking the works wherein C. tenuis was regarded as a junior synonym of C. lighti , we found that all also treated C. lighti and C. joyneri as valid species ( Wu 1932; Kamohara 1953; Zheng 1955; Ochiai 1963; Cheng & Weng 1965; Cheng 1997). According to our conclusion that C. lighti and C. joyneri are conspecific, only comparisons between C. joyneri and C. tenuis were explored further in our study.

Because the holotype of C. tenuis has been lost, we relied on meristic, morphometric, and descriptive information about morphological features of C. tenuis reported in the original description to compare with those of C. joyneri . Twelve features of C. tenuis from the original description ( Table 5 ) were compared with those of C. joyneri , including two characters, BD/SL and type of scales on both sides of the body. Both characters were regarded by Oshima (1927) as being diagnostic for distinguishing these two nominal species. All 12 characters of C. tenuis were the same as or included in the ranges of characters observed in C. joyneri ( Table 5 ). These results support the opinion previously proposed by Kamohara (1953), Ochiai (1963), Menon (1977) and others ( Lindberg & Fedorov 1993; Ho & Shao 2011; Kottelat 2013) that C. tenuis (Oshima) is a junior synonym of C. joyneri .

For C. tshusanensis, Chabanaud (1951) examined 12 morphological features ( Table 5 ) and pointed out that, although similar in many features, this species could be distinguished from C. joyneri and C. lighti by differences in three meristic characters (AFR, DMLL, and MLL) and six morphometric features (HL/SL, SNL/HL, LED/HL, IOW/HL, POL/HL and BD/SL). We examined the holotype of C. tshusanensis and compared 22 of its morphological features with those of C. joyneri and C. lighti ( Table 5 ). All of these morphological characters overlapped completely or were very similar between C. tshusanensis and C. joyneri ( Table 5 ), except for the count of MLL. We obtained an MLL count for the holotype of C. tshusanensis of 77 scales, which is lower than that (82–85 and 90–92) reported by Chabanaud (1951) in his study of the type specimens of C. joyneri and C. lighti , respectively ( Table 5 ), but within the range of MLL counts (62–80) we observed for specimens of C. joyneri . We regard these differences in MLL counts to be the result of different methods for counting these scales between our study and that employed by Chabanaud (1951). We conclude that these differences do not reflect interspecific differences between nominal species, instead, based on the overall similarity between these nominal species, our results support the view expressed in previous works ( Menon 1977; Lindberg & Fedorov 1993; Desoutter et al. 2001; Kottelat 2013; Fricke et al. 2017) that C. tshusanensis is a junior synonym of C. joyneri .

Figures

FIGURE 1. Photographs of ocular sides of lectotypes of two nominal species of Cynoglossus.A) Cynoglossus joyneri (BMNH 1878.4.15.94, 232.6 mm SL); and B) Cynoglossus lighti (BMNH 1924.12.15.87, 172.0 mm SL). Photos by K. Webb (Natural History Museum, London).

FIGURE 2. Twenty-twosampling locations for 138 specimensof tongue solesexamined in thisstudy.Abbreviationsfor locations are: NJ: Niphon; SJ: Shizuoka; KJ: Kochi; QH: Qinhuangdao; YT: Yantai; QD: Qingdao; RZ: Rizhao; LY: Lianyungang; LS: Lvsi; SH: Shanghai; ZS: Zhoushan; WL: Wenling; WZ: Wenzhou; FZ: Fuzhou; MW: Mawei; XM: Xiamen; DS: Dongshandao; SW: Shanwei; SZ: Shenzhen; ZH: Zhuhai; JM: Jiangmen; BM: Baimajing.

FIGURE 4. Close-up photographs of heads of lectotypes of: A) Cynoglossus joyneri; and B) Cynoglossus lighti. Preorbital lateral lines indicated by arrows. Photos by J. Maclaine and K. Webb (Natural History Museum, London).

FIGURE 5. Photographs of specimens representing the three groups of tongue soles identified in this study. A) First group featuring wider body and shorter rostral hook. B) Second group featuring narrower body and longer rostral hook. C) Third group: C-1) featuring wider body and longer rostral hook; C-2) featuring narrower body and shorter rostral hook. Photos: A and C-1 by H.-R. Luo; B and C-2 by X.-Y. Kong.