Sardinella gibbosa ( Bleeker, 1849 )

Sardinella gibbosa ( Bleeker, 1849) View in CoL

[English name: Goldstripe Sardinella ; new standard Japanese name: Nankai-sappa] ( Fig. 1 View Fig ; Tables 1, 2)

Material examined. 18 specimens, 111.2–142.4 mm SL: URM-P 6388, 142.4 mm SL, Nakagusuku Bay , Okinawa Island , Ryukyu Islands , Japan, 28 December 1982, set net; URM-P 8673, 137.7 mm SL, URM-P 8674, 137.1 mm SL, URM-P 8675, 116.3 mm SL, Nakagusuku Bay , 25 April 1984, set net; URM-P 32851, 137.8 mm SL, Tsuken Island, located within Nakagusuku Bay, 23 March 1990, set net; URM-P 44674, 128.9 mm SL, URM-P 44675, 117.8 mm SL, URM-P 44676 , 124.0 mm SL, URM-P 44677, 118.4 mm SL, URM-P

44678, 117.0 mm SL, URM-P 44679 , 131.0 mm SL, URM-P 44680, 128.3 mm SL, URM-P 44681, 124.6 mm SL, URM-P 44682 , 130.0 mm SL, URM-P 44683, 127.3 mm SL, off Umino Fishing Port , Nakagusuku Bay, 11 November 2008, set net, N . Miura; URM-P 44692, 132.6 mm SL, off Umino Fishing Port, Nakagusuku Bay , 11 November 2008, set net, N . Miura; URM-P 44697, 111.2 mm SL, URM-P 44698 , 130.0 mm SL, off Umino Fishing Port , Nakagusuku Bay, 20 November 2008, set net, H . Ishimori.

Description. Counts and measurements given in Tables 1 and 2, respectively. Body rather elongate, laterally compressed, deepest at dorsal-fin origin. Dorsal profile gently elevated from snout tip to dorsal-fin origin, thereafter gradually lowering to uppermost point of caudal-fin base. Ventral profile lowering from lower-jaw tip to just below pectoral fin, thereafter nearly straight to anus, before gently rising to lowermost point of caudal-fin base. Ventral edge of body covered with 32–34 keeled scutes from isthmus to anus. Pectoral-fin insertion slightly anterior to posteriormost point of opercle, below level of snout tip. Dorsal, ventral, and posterior margins of pectoral fin nearly straight; posterior tip of fin pointed, not reaching pelvic-fin insertion. Uppermost fin ray of pectoral fin not extended as filament. Pelvic-fin insertion below base of seventh to tenth dorsal-fin ray. Posterior tip of depressed pelvic fin not reaching anus. Dorsal-fin origin posterior to posterior tip of pectoral fin. Anterior margin of dorsal fin elevated from fin origin to fourth ray tip; middle portion of dorsal-fin margin slightly concave. Posteriormost dorsal-fin ray not filamentous. Anus situated just anterior to anal-fin origin. Anal-fin origin posterior to posterior end of dorsal-fin base. Two posteriormost anal-fin rays enlarged. Caudal fin forked, posterior tips of both lobes pointed. Dorsal and ventral margins of both lobes of caudal fin nearly straight. Snout tip pointed, slightly behind tip of lower jaw. Eye large, round, covered with adipose eyelid, positioned laterally on head dorsal to horizontal through pectoral-fin insertion, visible in dorsal view. Pupil round. Interorbital space flat. Orbit elliptical. Nostrils close to each other, anterior to orbit. Mouth terminal, small, posterior tip of maxilla slightly short of or just below anterior margin of pupil. Ventral margin of maxilla with small uniserial teeth. Premaxilla and hypomaxilla without teeth. First supramaxilla elongate. Second supramaxilla paddle-shaped, symmetrical. Posterior ramus of lower jaw elevated. Posterior margins of preopercle, subopercle and opercle convex, rounded, without serrations. Pseudobranchial filaments present, exposed. Two fleshy outgrowths on posterior margins on gill opening; a single large papilla on ventral margin. Gill rakers long, slender, with numerous asperities on anterior and posterior faces. Scales cycloid, thin, decidu- ous, those on lateral body surface with several centrally discontinuous vertical striae, posteriorly somewhat fimbriated ( Fig. 2 View Fig ). Predorsal scales paired. No elongate wing-like scales present beneath normal paired scales. No scales on head and most fins; a broad triangular sheath of scales on caudal fin.

Color of preserved specimens: Dorsum to upper part of lateral surface of body dark brown. Lower part of lateral surface and ventral surface of body uniformly silver or pale brown. Several narrow longitudinal stripes on upper part of body. Pectoral, pelvic, and anal fins semitransparent, pale, without melanophores except for uppermost pectoral-fin ray. Dorsal and caudal fins uniformly pale, melanophores scattered along fin rays. Dark blotch on dorsal-fin origin ( Fig. 1J View Fig ). Melanophores densely scattered on tips of both jaws.

Distribution. Sardinella gibbosa is widely distributed in the Indo-West Pacific from the east coast of Africa (including the eastern Mediterranean Sea as invasive species) to northern Australia and the Ryukyu Islands ( Whitehead 1985; Munroe et al. 1999; Matsunuma 2011, 2013, 2018; Miura 2012; Stern et al. 2015, 2016; Hata 2017a, 2019; this study). In Japan, the species is known only from Nakagusuku Bay, southeastern Okinawa Island, Ryukyu Islands ( Miura 2012; this study).

Identification. The specimens collected from Okinawa Island were assigned to the genus Sardinella as defined by Whitehead (1985) and Munroe et al. (1999), having the abdomen covered with prominently keeled scutes, paired predorsal scales, a symmetrical second supramaxilla, toothless hypo-maxilla, two posteriormost anal-fin rays enlarged, the dorsal fin without filamentous rays, and two fleshy outgrowths on the hind margin of the gill opening. Moreover, they conformed to Sa. gibbosa , having the following combination of characters that closely matched the diagnostic features given by Whitehead (1985), Munroe et al. (1999), and Stern et al. (2016): caudal fin uniformly pale, without distinct blotch on posterior tips of both lobes; black spot on dorsal-fin origin; striae on scales on lateral surface of body interrupted in the center; 26−31+50–57=77–88 gill rakers on first gill arch; and 18 or 19+14 or 15=32–34 keeled scutes along the body ventral surface. In particular, their meristic and morphometric characters generally matched those given by Stern et al. (2016) for western Pacific Sa. gibbosa , although they slightly differed from Stern et al.’s (2016) specimens in having slightly higher total ray counts for the anal and pectoral fins, generally more pseudobranchial filaments (18–21, 15–16, and 17–23 vs. 17–20, 14–15, and 14–21), and a slightly narrower body (22.8–26.2% SL vs. 23.8–29.8%). Because ranges of anal- and pectoral-fin ray numbers greater than five and three, respectively, have been reported in congeneric species ( Stern et al. 2016; Hata and Motomura 2019a –d), the pseudobranchial filament and body depth differences are considered as interspecific variations only.

Although Sa. gibbosa closely resembles Sardinella goni Stern, Rinkevich, and Goren, 2016 (recorded only from Boracay Island, Philippines), the two species sharing body scales with discontinuous vertical striae, a black spot on the dorsal-fin origin and caudal fin without black blotch, and almost identical numbers of gill rakers on the first gill arch, the former has 14 or 15 postpelvic scutes (vs. 13 in S. goni ; Stern et al. 2016).

Taxonomic status of Clupea immaculata Kishinouye, 1908 . Sardinella gibbosa was described by Bleeker (1849) as Clupea gibbosa based on one or more specimens collected from Makassar, southwestern Sulawesi, Indonesia. Shortly after the original description of C. gibbosa, Bleeker (1851) described Spratella tembang as a new species, without any descriptions, and listed under C. gibbosa as a synonym despite the reason for the replacement is not clear ( Whitehead et al. 1966). Spratella tembang is considered as an unneeded replacement name for C. gibbosa and the two nominal species must share the same type specimens ( Whitehead et al. 1966; ICZN 1999: Article 72.7 of the International Code of Zoological Nomenclature, hereinafter the Code). Whitehead et al. (1966) pointed that Bleeker (1849) gave no size range or number of specimens, which or how many specimens belonged to the type series of the nominal species being uncertain. They indicated BMNH 1867.11.28.46, a specimen collected from Makassar, as “putative neotype ” of C. gibbosa Bleeker, 1849 . Subsequently, Stern et al. (2016), treated the BMNH specimen as the neotype of the nominal species. However, a formal designation was not provided by Whitehead et al. (1966) or Stern et al. (2016); therefore, the neotype designation failed to meet the full mandatory requirements listed under Article 75.3.1 of the Code. The formal neotype designation is needed if taxonomic confusion related to Sa. gibbosa is found in the future.

Whitehead (1985), who reviewed the genus Sardinella , considered Clupanodon jussieu Lacepède, 1803 , Spratella tembang Bleeker, 1851 , Clupea immaculata Kishinouye, 1908 , Fimbriclupea dactylolepis Whitley, 1940 , and Sardinella taiwanensis Raja and Hiyama, 1969 to all be junior synonyms of Sa. gibbosa . Subsequently, Stern et al. (2015, 2016) regarded Harengula dollfusi Chabanaud, 1933 , believed to be a junior synonym of Sardinella albella (Valenciennes, 1847) by Whitehead (1985), as a junior synonym of Sa. gibbosa , based on examination of the syntypes of H. dollufusi . In addition, Hata and Motomura (2019d) regarded Clupa. jussieu as a nomen dubium, leaving Sp. tembang , C. immaculata , H. dollfusi , F. dactylolepis , and Sa. taiwanensis as junior synonyms of Sa. gibbosa .

One of these, C. immaculata , was described by Kishinouye (1908). Although Whitehead (1985) regarded C. immaculata as a junior synonym of Sa. gibbosa , this was questioned by Aonuma and Yagishita (2013). Although Kishinouye (1908) stated that he collected one specimen each from Saga, Japan, Swatow, Guangdong, China, and Xiamen, Fujian, China, his description of C. immaculata included meristics without value ranges, suggesting that the description was based only on a single specimen. However, it is unknown which specimen the description was based on. Unfortunately, the type specimens of C. immaculata have since been lost ( Wongratana 1980; Fricke et al. 2021; this study). In fact, Kishinouye’s (1908) characters of C. immaculata generally matched those of Sa. gibbosa shown in Whitehead (1985), Munroe et al. (1999), and Stern et al. (2015, 2016), as well as the Okinawan specimens examined here: dorsal- fin rays 17, pectoral-fin rays 15, pelvic-fin rays 8, abdomen covered with 19+14 scutes, and dorsal-fin origin nearer to snout tip than caudal-fin base. However, Sa. gibbosa has at no time been recorded from temperate Japanese waters, including Saga (northern coast of Kyushu), one of the type localities of C. immaculata although intensive ichthyofaunal surveys have been carried out in Japanese waters ( Aonuma and Yagishita 2013; Motomura 2020).

On the other hand, the abovementioned characters of C. immaculata also match those of Sardinella aurita Valenciennes, 1847 , except the pelvic-fin ray count [8 in Kishinouye (1908) ’s C. immaculata vs. 9 in Sa. aurita ( Whitehead 1985; Munroe et al. 1999; Stern et al. 2017)]. Furthermore, Sa. aurita has been frequently reported from southern Japan, including the northern coast of Kyushu (frequently reported as Sa. lemuru Bleeker, 1853 ) [e.g., Omori 2007; Kagoshima City Aquarium Foundation 2008, 2018; Hasegawa 2011; Hata and Motomura 2011, 2017; Kadowaki et al. 2015; Kaburagi 2016; Tashiro et al. 2017; Hata 2017b, 2018a, b, 2020; Funaki and Saitoh 2018; Kobayashi 2019; Nakashimada and Hibino 2020; Hata and Koeda 2020: fig. 2]. Judging from these facts, Kishinouye’s (1908) specimen from Saga was probably Sa. aurita . To confirm the taxonomic status of C. immaculata , more researches are needed.

Previous records of Sa. gibbosa from Japan. As mentioned above, a number of unidentified species of Sardinella has been reported from Japan. In addition, C. exile Kishinouye, 1911 was described as a new species, based on specimens collected from Chichi-jima Island, Ogasawara Islands, Japan. In the original description of the species, Kishinouye (1911) stated that it was the most abundant clupeoid fish in the Ogasawara Islands. As with C. immaculata , no type specimens of C. exile exist ( Wongratana 1980; Fricke et al. 2021; this study). Subsequently, Matsubara (1955) treated C. exile as a junior synonym of Sardinella jussieu (Lacepède, 1803) , a nominal species regarded as nomen dubium by Hata and Motomura (2019d). Clupea exile was not treated by Whitehead (1985) in his taxonomic review of the family Clupeidae and the taxonomic status of the nominal species is unknown. Aonuma and Yagishita (2013) similarly excluded C. exile in their review of Japanese clupeoids, due to the lack of detailed information on the species. However, the vertebral count of C. exile described by Kishinouye (1911) was significantly lower than in specimens of Sa. gibbosa examined here (40 vs. 46–49; Table 1). Therefore, C. exile is clearly not synonymous with Sa. gibbosa . In fact, Yoshigou (2002) suggested that C. exile is a junior synonym of Herklotsichthys quadrimaculatus (Rüppell, 1837) , based on the correspondence of characters of C. exile , shown by Kishinouye (1911), and specimens of H. quadrimaculatus collected from the Ogasawara Islands, reported by Yoshigou (2002). To clarify the taxonomic status of C. exile , further consideration of closely related genera is necessary.

Aoyagi (1941) reported three clupeoid specimens (54.7– 60.2 mm SL) collected from Miyako Island, Japan, as Sardinella sindensis (Day, 1878) , although the species is considered to be distributed only in the northwestern Indian Ocean, from the Gulf of Aden to the western coast of India ( Whitehead 1985). Subsequently, Matsubara (1955) report- ed the following diagnostic characters of Sa. sindensis sensu Aoyagi (1941) : pelvic fin with eight fin rays, black spot on dorsal-fin origin, lower gill rakers on first gill arch 58–62, body depth 3 to 4 times (approx. 25–33.3%) in SL, and eye diameter 3.5 to 4 times (approx. 25–28.6%) in head length. Although the pelvic-fin ray count, dorsal-fin marking, and eye diameter to head length ratio of Sa. sindensis given by Matsubara (1955) closely matched those of Okinawa specimens of Sa. gibbosa presented here, the gill-raker counts and body depth in SL ratio were quite different. Accordingly, Sa. sindensis sensu Aoyagi (1941) is not considered to be Sa. gibbosa , although its possible identity as Sa. albella or Sa. alcyone cannot be discounted, due to their similar lower gill-raker counts and deeper body (counts of lower gill rakers on first gill arch 47–64 in Sa. albella in 63.9–130.0 mm SL specimens, 67–72 in Sa. alcyone in 66.6–109.8 mm SL specimens; body depth 28.8–33.8% SL in Sa. albella , 26.4– 36.8% in Sa. alcyone ) ( Stern et al. 2016; Hata and Motomura 2019c). It should be noted, however, that the northern distribution limit of Sa. albella in the Pacific Ocean is Taiwan, the species not having been recorded from Japanese waters ( Whitehead 1985; Munroe et al. 1999; Stern et al. 2016; Hata 2019).

Miura (2012) reported an unidentified photograph of a sardine from Nakagusuku Bay, southern coast of Okinawa Island, where specimens of Sa. gibbosa examined in the present study were collected, as “Sappa-zoku-no-isshu-2” (meaning Sardinella sp. 2 in Japanese), noting that the species was rarely caught in Nakagusuku Bay. Although the specimen photographed is here presumably identified as Sa. gibbosa due to its elongated body, this identification cannot be validated as no voucher specimens have been retained in Miura (2012). Consequently, the specimens described herein represent the first records of Sa. gibbosa from Japanese waters, with Nakagusuku Bay, Okinawa Island being confirmed as the northern distribution limit of the species.

Because no Japanese name has previously been applied to specimens of Sa. gibbosa , the new standard Japanese name “Nankai-sappa” is herein proposed for the species (based on URM-P 44692; Figs 1F View Fig , 2 View Fig ), “nankai” meaning “southern sea”, in reference to the primarily tropical distribution of the species, and “sappa” being the Japanese name for the genus Sardinella .

Comparative material examined. BMNH 1867.11 . 28.46, Makassar , Indonesia; listed as “putative neotype of Clupea gibbosa ” in Whitehead et al. (1966) (see “Taxonomic status of Clupea immaculata Kishinouye, 1908 ”).