Larval study revealed diversity and life-history traits of crypto-benthic eel gobies.

Pelagic larval duration and settlement sizes in species of Apogonidae, Labridae, Scaridae, and Tripterygiidae in a coral lagoon in southern Japan were examined. Sampling was conducted monthly from July 2004 to June 2005 in the coral lagoon and channel of the Oh-do Beach on Okinawa Island, Japan. Pelagic larval duration was estimated by the number of otolith increments. Mean standard length at settlement of apogonids ranged from 7.7 to 13.9 mm, and mean pelagic larval duration ranged from 14.0 to 30.6 days (14 species, 418 individuals). In labrids, mean standard length at settlement and pelagic larval duration varied greatly (mean standard length: 5.4–11.0 mm; pelagic larval duration: 18–57 days, four species, four individuals). Scarids showed consistent mean standard length at settlement and pelagic larval duration (mean standard length: 7.1–7.6 mm; pelagic larval duration: 29–42 days, five species, 25 individuals). In tripterygiids, pelagic larval duration was more consistent (range: 18–29 days, mean: 22.2 ± 2.1 days), but mean standard length at settlement ranged from 7.8 to 10.3 mm (six species, 32 individuals). These results suggest that the pelagic larval duration of Apogonidae and Tripterygiidae (nonpelagic egg spawning) is shorter than that of Labridae and Scaridae (pelagic egg spawning), and the dispersal strategy of labrids and scarids may include wider dispersal than that of apogonids and tripterygiids.

Recruitment courses of three amphidromous sleeper species, Eleotris acanthopoma, E. melanosoma, and E. fusca, were investigated at the surf zone adjacent to the river mouth and at five stations in the Teima River on Okinawa Island, Japan. All three species occurred at the surf zone as pelagic larvae with transparent and compressed body, a conspicuous air bladder, and an emarginated caudal fin. Eleotris fusca (16.0–19.6 mm in standard length: SL) sometimes possessed a vestige of the larval chin barbel and were larger than E. acanthopoma (9.7–13.2 mm SL) and E. melanosoma (11.2–12.8 mm SL). The pelagic larvae were also collected during full tide from the lower reaches of the tidally influenced area of the river. The pelagic larvae may be carried in and out of the estuary with some tidal fluxes, and they may settle when they reach the upper tidally influenced area where the salinity becomes extremely low. Body width and pigmentation of newly settled larvae increased. E. fusca was considered to migrate upstream to the freshwater area against the flow of the river just after reaching the settled stage. After settlement, all three species became completely pigmented, the caudal fin became round in shape, and the fin ray counts became complete with growth. Also, E. acanthopoma dispersed widely to the lower part of the tidally influenced area or to the lower reaches of the freshwater area, E. melanosoma dispersed to the lower part of the tidally influenced area, and E. fusca dispersed upstream.

Larvae and juveniles of the alepocephalid fishes, Leptoderma lubricum [26.9–69.0 mm in standard length (SL)] and Leptoderma retropinnum (21.1–67.2 mm SL), collected within 1–8 m of the seafloor in Suruga Bay, southern Japan, are described. They can be easily distinguished from each other by the following adult-like characters: membrane morphology between the vertical fin rays and procurrent caudal-fin rays (separated in L. lubricum vs. continuous in L. retropinnum), numbers of dorsal-fin rays (34–40 vs. 45–52) and anal-fin rays (50–57 vs. 65–72), and caudal peduncle length (11.7–13.4% SL vs. 4.5–5.9% SL), in addition to several other body proportional differences. Unique characters in the larval stages of Leptoderma include a translucent occipital region, horizontally elongated eye, and head below the upper margin of the orbit and abdominal cavity densely covered by melanophores, ontogeny being characterized by the acquisition of general adult characters to the postflexion stage, indistinct transformation, and the retention of few larval characters until almost the end of the juvenile stage, as in other known alepocephalids. In addition to the near-bottom larval and juvenile collections of both species, the occurrences of benthic or near-bottom taxa, including Harpacticoida, in their gut contents confirmed the early life history dependence of the former on the near-bottom.

Larvae and juveniles of six species of Luciogobius were collected at Aritsu Beach on Okinawa Island using a small seine. Postflexion larvae were dominant during sampling and were collected when they approached the shoreline adjacent to or at the entrances to their adult habitats prior to settlement. Standard lengths of postflexion larvae ranged from 5.4 to 14.4 mm and varied depending on the species. The larvae occurred mainly from January to April, but some larvae were caught in October and November. Their pelagic larval durations were estimated to range from 17 to 36 days and varied depending on the species. Morphologies of field-caught larvae and juveniles and laboratory-reared juveniles were described. Six species were clearly distinguished based on fin ray and vertebral counts, proportions, body size, and pigment patterns. Although their taxonomic statuses could not be determined, it is thought that they have independent relatives in other regions.

Reproductive biology and early development of two species of sleepers, Eleotris acanthopoma Bleeker, 1853, and E. fusca (Forster, 1801), were investigated in streams on Okinawa Island in southern Japan. Gonadal examination and morphology of the genital papillae indicated that E. acanthopoma matured at a smaller body size (ca. 28 mm in standard length) than E. fusca (ca. 50 mm). Mature ovaries were composed of oocytes that could be categorized into two size classes. Larger females of both species had several hundred thousand developed oocytes in the larger size class and may spawn them at one or several consecutive spawning events. Egg masses of both species were found in habitats typically occupied by adults and were deposited, often sparsely, on the underside of objects. Form of the egg masses and morphology of eggs and newly hatched larvae of both species were almost identical. Eggs were a nearly spherical pyriform in shape, with the widest diameters measuring approximately 0.4 mm. Newly hatched larvae were very small (1.0–1.4 mm in notochord length) and undeveloped. The mouth opened and the eyes became pigmented 3 days after hatching, and all of their yolk was consumed 4 days after hatching. Reproductive strategies of both species were characterized by high fecundity through production of small eggs and small newly hatched larvae, with high fecundity likely to mitigate the presumed increased risk associated with widespread larval dispersal.

Age, growth, and reproductive biology of the Waigieu seaperch Psammoperca waigiensis were studied using 291 specimens obtained around Okinawa Island, Japan. Otolith opaque zones that formed every year correlated with spawning activity and were thought to be annual rings. Growth of this species was rapid during the first 2 years, reaching 186.2–270.3 mm in standard length (SL). Females (196.6–334.0 mm SL) were larger than males (186.2–288.6 mm SL), caused by differential growth between sexes, which started before 2 years of age. Most of the specimens were 1–11 years old and accounted for 96% in total. Spawning season was estimated to be from April to October by gonadosomatic index (GSI) and histological observation. The smallest mature female and male were 217.0 mm SL (2 years) and 206.0 mm SL (2 years), respectively. After recruitment in rocky areas up to about 200.0 mm SL and 2 years of age, Psammoperca waigiensis were then found to soon mature.

The reproductive cycle and spawning characteristics of the Japanese gizzard shad (Nematalosa japonica) were examined histologically using specimens collected around Okinawa Island in southwestern Japan. The observed lengths at 50 % maturity were approximately 170 mm in standard length (SL) for females and 160 mm SL for males, which corresponded approximately to 3-year-old fish. The spawning season was estimated to take place from January to May, peaking from February to April, which closely coincided with the coldest water temperatures of the year. The lipidosomatic index of both sexes decreased from January onward and maintained low values until July, presumably as the fish expended energy to reproduce. Spawning intervals were estimated to occur between 1.3 and 1.8 days from February to April. Fecundity (F) for females tested using 69 ovaries was related to SL: F = 1.313 × 10−4 × SL3.966. Histological observations and a fish market survey suggested that the species spawned in sandy muddy areas (<20 m) based on the presence of mature individuals. The implications of our results toward conservation and management of N. japonica are discussed.

Thirteen cohorts of the Caribbean reef fish Thalassoma bifasciatum were collected over 4 yr (2000 to 2003) in the upper Florida Keys, USA. Juvenile fish were censused and collected from replicate reefs shortly after settlement. The otoliths were examined to obtain early life history information such as timing of spawning, larval growth, pelagic larval duration, size-at-age, timing of set- tlement, and juvenile age and growth. Mean water tem- perature over the reef during the larval period explained 78% of the variation in larval growth among cohorts. Faster-growing warm-water fish had shorter pelagic lar- val durations (PLD), and larval growth explained 85% of the variation in PLD. Relative (otolith) size-at-settlement was a function of larval growth as well as PLD: settling larvae were largest at intermediate water temperatures. Early juvenile growth was also directly related to water temperature, which enabled smaller warm-water settlers to grow rapidly and eventually exceed juvenile size-at- age of the cooler-water cohorts. Cohorts encountering intermediate water temperatures remained the largest throughout early juvenile life on the reef. The relative size of recruitment events could not be explained by any larval or juvenile trait, nor by water temperature. Recruitment was generally low for cool-water cohorts but quite vari- able among warm-water cohorts. This may be due to in- creased difficulty in sustaining high growth rates in warm water (i.e. beyond a threshold temperature of 28.5°C) or the interference of mesoscale advection processes. When 4 cohorts that settled during the passage of mesoscale eddies were omitted from the analysis, 61% of the varia- tion in recruitment magnitude could be explained by water temperature alone. The dynamic oceanographic setting of the Florida Keys may obscure the relationship between seasonal water temperature, early life history traits, and magnitude of recruitment events.

The Western Indian Ocean harbors one of the world’s most diverse marine biota yet is threatened by exploitation with few conservation measures in place. Primary candidates for conservation in the region are the Scattered Islands (Iles Eparses), a group of relatively pristine and uninhabited islands in the Mozambique Channel. However, while optimal conservation strategies depend on the degree of population connectivity among spatially isolated habitats, very few studies have been conducted in the area. Here, we use highly variable microsatellite markers from two damselfishes (Amphiprion akallopisos and Dascyllus trimaculatus) with differing life history traits [pelagic larval duration (PLD), adult habitat] to compare genetic structure and connectivity among these islands using classic population structure indices as well as Bayesian clustering methods. All classical fixation indexes F ST, R ST, G′ST, and Jost’s D show stronger genetic differentiation among islands for A. akallopisos compared to D. trimaculatus, consistent with the former species’ shorter PLD and stronger adult site attachment, which may restrict larval dispersal potential. In agreement with these results, the Bayesian analysis revealed clear genetic differentiation among the islands in A. akallopisos, separating the southern group (Bassas da India and Europa) from the center (Juan de Nova) and northern (Iles Glorieuses) islands, but not for D. trimaculatus. Local oceanographic patterns such as eddies that occur along the Mozambique Channel appear to parallel the results reported for A. akallopisos, but such features seem to have little effect on the genetic differentiation of D. trimaculatus. The contrasting patterns of genetic differentiation between species within the same family highlight the importance of accounting for diverse life history traits when assessing community-wide connectivity, an increasingly common consideration in conservation planning.

The gerreid species,Gerres baconensis (Evermann & Seale),G. equulus Temminck & Schlegel andG. oyena (Forsskal), were re-assessed as valid following examination of their holotypes and other specimens, and included in the “G. oyena complex”.Gerres haconensis is currently known only from Bacon, Luzon Island, Philippines and the Ogasawara (=Bonin) Islands, Japan, andG. equulus only from southern Japan (except Ryukyu Islands) and southern Korea.Gerres oyena is widely distributed in the Indo-West Pacific (in Japan, only from the Ryukyu Islands).Gerres baconensis differs from bothG. equulus andG. oyena in having higher counts of both the pored lateral line scales (39–42 vs 35–41 in the latter two species) and the lower gill raker series (8 or 9 vs. usually 7). A U-shaped premaxillary groove, formed on the dorsum of the forehead by the long ascending processes of the premaxillae, is scaleless inG. equulus andG. oyena, whereas it is fully scaled just behind the level of the posterior nostrils inG. baconensis over ca. 160 mm in standard length (SL) (partially scaled in specimens of ca. 100 mm SL).Gerres equulus differs fromG. oyena in having the posterior margin of the maxilla not extending beyond a vertical through the anterior margin of the inner dermal eye opening, shorter second dorsal and anal fin spines (means 18.5% and 8.5% of SL, respectively vs. 21.2% and 10.3% of SL), lower body depth at first anal fin spine base (27.0% vs. 29.6% of SL) and dorsomedial U-shaped groove scaleles throughout life (vs. tiny squamation anteriorly in specimens over ca. 130 mm SL). OtherGerres species of uncertain status and/or related species are discussed.

The blackspot snapper Lutjanus fulviflammus is one of the most common Lutjanus species landed and consumed on Okinawa Island, Japan. Using 901 fish (29.9–304.2 mm in standard length; SL) obtained from fish markets or caught by angling, the age, growth, maturation size and maturation age were estimated. Each opaque zone formed on the otolith every year correlated with their spawning and was thought to be an annual ring. The main spawning season was estimated to be from April to July, which peaked in May and June, with a few mature fish collected in August and September. Maximum ages were observed as 24 years for both sexes and L∞ (mm SL), K and t0 (years) of the von Bertalanffy growth formula were estimated to be 276, 0.144 and −5.22 for females and 247, 0.227 and −3.18 for males, respectively. First maturation size and age were estimated to be 175 mm SL and 2 years for both sexes. A faster growth rate up to 2 years will allow reproduction at a young age and provide many chances for spawning throughout their long life. Furthermore, the possibility exists that populations would easily recover if regulations were established.

Age, growth, and reproductive characteristics of blackspot snapper Lutjanus fulviflammus around the Yaeyama Islands (24°N, 124°E) in southern Japan were investigated between 2010 and 2014. Samples were caught by angling or purchased monthly at fish markets to cover all size ranges and seasons. Specimens consisted of 280 females (176–347 mm fork length), 177 males (193–325 mm), and 17 juveniles (42–128 mm). Spawning was confirmed from April to August by the presence of postovulatory follicles, which, based on the occurrence of yolked oocytes possibly also occurred in March and September. Maximum fork length and age in the specimens around Yaeyama Islands were 347 mm and 23 years for females and 325 mm and 21 years for males. These results were compared to those reported previously for the species in other regions by standardized criteria. Possible spawning duration of 7 months in the population around Yaeyama Islands is similar to that in New Caledonia (22°S, 5 months) and Okinawa Island (26°N, 6 months), but shorter than that in Tanzania (8°S) and the southern Arabian Gulf (25°N) (~12 months). The maximum body sizes and ages around Yaeyama Islands are also similar to the populations in New Caledonia and Okinawa Island (>300 mm FL, >20 years old), but larger/older than those in Tanzania and the southern Arabian Gulf (<300 mm FL, <20 years old). These differences are thought to reflect water temperature, i.e. spawning duration is shorter in lower temperature regions (annual mean <26°C, New Caledonia, Okinawa Island, Yaeyama Islands) than in higher temperature regions (annual mean >27°C, Tanzania, southern Arabian Gulf). Subsequently, a shorter spawning duration extends the life-span as well as the maximum body size. Longevity of blackspot snapper is thought to be negatively correlated with the duration of the spawning season.

Thirty-eight museum specimens of the Ryukyu spiny rat, belonging to the genus Tokudaia Kuroda, 1943 (Rodentia, Muridae), from the islands of Amami Oshima and Okinawa, in the Ryukyu Islands (=the Nansei Islands), southern Japan, were examined and measured. Each specimen was classified into one of five age groups (I-V) determined by the wear of the three upper molars. The spiny rat of Okinawa has on average within age groups III and IV a statistically longer and wider skull and a longer molar row than that of Amami Oshima. However, neither the posterior nor the central parts of the skull length and width differ between the two islands. Against the same size of head and body length (H&BL) or incisor - the third upper molar length (I-M3), the spiny rat of Okinawa has longer I-M3 or narrower zygomatic arches than the spiny rat of Amami Oshima. The Okinawan spiny rat has a wider first upper molar (wM1≥1.9 mm), whereas that of Amami Oshima is wM1≤1.8 mm. The Okinawan spiny rat has the palatine foramen of the skull situated more posteriorly than that of Amami Oshima. These findings indicate conclusively that the spiny rat populations on the two islands of Amami Oshima and Okinawa have distinctly different morphological characteristics, in addition to having different karyotypes, and as such they should be regarded as distinct species. The spiny rat of Amami Oshima should be named as Tokudaia osimensis (Abe, 1933) while that of Okinawa should be known as Tokudaia muenninki (Johnson, 1946).

Callogobius falx n. sp. (Gobiidae) is described on the basis of two specimens (14.4 and 23.1 mm in standard length: SL) from the Ryukyu Islands, southern Japan. The new species is characterized by the following combination of characters: dorsal-fin rays VI-I, 8 or 9; anal-fin rays I, 7; pectoral-fin rays 18 or 19; lateral scale rows 2022; body depth moderate, neither particularly slender nor deep (depth at pelvic-fin origin 18.320.0% of SL); pelvic-fin base with frenum; caudal-fin length moderate, margin rounded; cephalic sensory-canal pores absent; interorbital papillae row separated, forming two rows (one row in 14.4 mm paratype); postorbital papillae row not continuous with upper cranial papillae row; preopercular and transverse opercular papillae rows connected, forming single transverse row; body with three oblique black bars, angled upwards anteriorly and continuous with black bars on dorsal fins, anteriormost bar extending from 1st dorsal fin to midline of body, posterior two bars extending from second dorsal fin to dorsal surface of body and caudal peduncle, respectively; pectoral fin white, upper half widely (ca. 1/2 length of rays) margined with black; caudal-fin base with narrow sickle-like bar.

Larva and juveniles belonging to two ophidiid species of the genus Brotula Cuvier 1829, Brotula sp. 1 [3 specimens, 14.2, 50.7, and 55.7 mm in standard length (SL)] and Brotula sp. 2 (1 specimen, 47.8 mm SL), were collected from the Northwest Pacific Ocean off the Okinawa, Izu (Aoga-shima Island and Miyake-jima Island), and Ogasawara Islands, Japan. Although the counts and proportions of both species agree with the documented ranges of Brotula multibarbata Temminck and Schlegel 1846 known from Japan, the larva and juveniles are clearly different from each other by the following combination of characters: counts of dorsal-fin rays [Brotula sp. 1, 122–126 (about 130 in the larva) vs. Brotula sp. 2, 118], anal-fin rays [94–101 (106) vs. 85], caudal-fin rays (11 vs. 9), and vertebrae (14 + 45–46 = 59–60 vs. 16 + 42 = 58); position of dorsal-fin origin [above or posterior to pectoral-fin base (above 6th vertebra) vs. anterior to pectoral-fin base (above 3rd vertebra)]; predorsal length (56.5–59.8% vs. 51.0% of preanal-fin length); blackish spots behind eye (present vs. absent); brown spots on body and vertical fins (present vs. absent); size of pectoral fin [small, not reaching anus (13.5–13.7% of SL) vs. large, reaching anus (28.3% of SL)]; and length between dorsal- and anal-fin origins (22.7–24.6% vs. 31.3% of SL). The occurrence of the larva and juveniles suggests that “B. multibarbata” known from Japan represents two species: Brotula sp. 1 and Brotula sp. 2 are probably B. multibarbata and Brotula formosae Jordan and Evermann 1902, respectively; the former is currently regarded as the senior synonym of the latter. In addition, this study describes for the first time the living coloration of Brotula larva.