Search

Aphyocypris chinensis is a small cyprinid that is widely distributed in lowland areas of continental China, the Korean Peninsula, and the northwestern part of Kyushu, Japan. However, this species is severely threatened in Japan due to human impacts; thus, several facilities and citizen groups implemented captive breeding and reintroduction/reinforcement programs using several dozen founder fish collected from Tanushimaru, the location with the only known natural population, in the 1980s and 1990s. To determine the phylogenetic position and genetic authenticity of Japanese populations of A. chinensis, we conducted mtDNA and genome-wide SNP analyses using a total of 254 specimens from 31 wild and captive samples, including specimens of Chinese and Korean A. chinensis and the close relative A. kikuchii from Taiwan. The mtDNA divergence and phylogeny indicated that the haplotypes found exclusively in Japan were differentiated from the Chinese/Korean haplotypes (uncorrected p = 2.6% in the cytochrome b gene) to the same extent as they were from A. kikuchii haplotypes (p = 2.9%). Results from mtDNA sequences and 47–359 SNPs, obtained using the MIG-seq method with different parameter sets, revealed that the initial captive populations and an extinct wild population in Ukiha, adjacent to Tanushimaru, were genuine Japanese populations, whereas all extant captive and wild populations in Japan are hybrids between Japanese A. chinensis and A. kikuchii, or A. kikuchii itself. The details of the captive breeding and exchange programs, as well as evidence for the aquarium trade of A. kikuchii since the 1990s, strongly suggest that the mixture of A. kikuchii into Japanese populations occurred within the first several years of the establishment of captive populations in 1994. The present case of the highly probable extinction of genuine Japanese populations of A. chinensis emphasizes the importance of confirmation and management of the genetic authenticity of conservation-targeted species.

The Chinese false gudgeon, Abbottina rivularis, is a common cyprinid fish that is widely distributed throughout continental East Asia, but exhibits a restricted, discontinuous distribution in western Japan, including Honshu and Kyushu islands. In this study, analyses of mitochondrial (cytochrome b) and nuclear (glyt, myh6, and RAG1) genes were conducted to investigate patterns and magnitudes of intraspecific differentiation among A. rivularis populations in Japan and adjacent continental areas. Phylogenetic analysis of the mitochondrial gene sequences resolved four major lineages—the Japan lineage (JL), a northern continental lineage (NCL), and two southern continental lineages (SCL1 and SCL2)—with uncorrected pairwise sequence distances of 9.4–15.2% (estimated divergence times, 7.9–17.1 Myr). Two lineages (JL and SCL1) occurred in both the Honshu and Kyushu districts of Japan. Compared with populations in continental areas, most Japanese populations exhibited less genetic diversity. The JL was divided into two well-differentiated sub-lineages distributed on Honshu and Kyushu islands, respectively. Kyushu Island, as well as areas on Honshu where the species is known to have been introduced, also harbored the SCL1 lineage, which constituted most of the populations on Kyushu. The applied nuclear DNA data strongly suggest that hybridization between the Japan and continental lineages has occurred on Kyushu Island. The artificial introduction hypothesis, instead of a two-origin scenario, best explains the origin of the SCL1 in Japan.

We evaluated the morphological, distributional, and genetic characteristics of the freshwater sculpin, Cottus pollux, in Kyushu Island, Japan. Based on pectoral fin ray number, the sculpins inhabiting 30 rivers were divided into two types [modes 13 (M13) and 15 (M15)]. We evaluated four environmental parameters: length of main river (L-MR), average gradient of main river (G-MR), distance from sampling site to river mouth (D-SM), and gradient around sampling site (G-S), and compared the two types. L-MR, D-SM, and G-S were significantly larger for the rivers that contained M13 fish than for those containing M15 fish. M13 individuals were distributed in the upstream areas of large river systems, suggesting a fluvial life history, whereas M15 individuals were distributed in the downstream areas of small river systems, indicating an amphidromous life history. We conducted phylogenetic analysis based on mitochondrial 12S rRNA [788 base pair (bp)] and control regions (386 bp). Mitochondrial DNA analysis showed that both M13 and M15 groups were genetically of C. pollux middle-egg type (ME). Our findings proposed the hypothesis that C. pollux ME in Kyushu Island exhibits dimorphism in both morphological and distributional traits. Additionally, haplotype distribution indicated that the fluvial M13 populations had higher genetic specificity in each river in contrast to the existence of one genetic group of amphidromous M15 individuals in Kyushu Island.

It has hitherto been unknown whether the paradise fish Macropodus opercularis that inhabits the Ryukyu Archipelago, Japan, is native to the region or was introduced. This study examined the genetic identity of fish from five islands in the Ryukyu Archipelago (Okinoerabu, Okinawa, Yagaji, Kume, and Minamidaito islands) and compared it with those from Taiwan Island, mainland China, and Hainan Island. Analyses of the mtDNA control (760 bp) and cytb (660 bp) regions showed that haplotypes of specimens from the Ryukyu Archipelago were the same as or were very similar (with a 1- or 2-bp difference) to those from Taiwan. In addition, haplotypes from the Ryukyu Archipelago also showed lower genetic diversity than those from Taiwan Island, mainland China, and Hainan Island. These results suggest a high likelihood that the fish in the Ryukyu Archipelago were artificially introduced from Taiwan. However, the possibility that the fish is indigenous to the Ryukyu Archipelago cannot be completely ruled out, because some haplotypes and a clade from the Ryukyu Archipelago have not been found in the other areas. Regardless of its origin, we emphasize the importance of the conservation of the paradise fish in the Ryukyu Archipelago as an indicator of the threatened wetland environment as well as for its cultural value.

We investigated the spawning nests of Gymnogobius macrognathos on a tidal flat in the Tatara River, Fukuoka Prefecture, Japan. Digging uncovered 19 spawning nests. The number of eggs and the standard length of the guarding male were positively correlated. Nine spawning nests were examined using in situ resin casting. All casts had structures characteristic of callianassid shrimp burrows and were most likely those of Nihonotrypaea japonica. Spawning nests had significantly greater average diameters than shrimp burrow openings and may have been widened by G. macrognathos.

Selecting candidate areas to protect the habitats of threatened freshwater fish species is a major challenge in urban areas. Therefore, we conducted a case study using a Marxan analysis based on fish fauna and land use data of five rivers flowing in the Fukuoka Plain, northern Kyushu Island, Japan, to construct priority maps to protect threatened freshwater fishes in the urbanized plain. We organized threatened fish fauna and land use data in the fourth-mesh scale, drawn as a 500-m square, resulting in 147 meshes with fauna and land use data. We assessed two cases—meshes were either analyzed as having the same cost (case 1) or having cost sizes classified into four classes based on cluster analysis of land use data (case 2). We divided the threatened species into those from either stream fish or floodplain fish and analyzed them separately. When meshes with high priority in both case 1 and case 2 were defined as primacy meshes for habitat protection, 10 and 28 meshes were identified in the stream and floodplain fish, respectively. Of the primacy meshes, 40 % for the stream fish and 54 % for the floodplain fish were distributed in urbanized areas, indicating that habitat protection was necessary in several urbanized areas on the plain. Meshes with low priority in case 1 and with high priority in case 2 were defined as the second most important meshes for habitat protection. The number of second most important meshes for the floodplain fish was smaller than that for the stream fish; therefore, the floodplain fish required habitat protection in urbanized areas to a greater extent than the stream fish.

To reveal the primary historical events that affected the geographic differentiation in freshwater organisms in the Japanese archipelago, we investigated the phylogeny and phylogeographic structure of the widely distributed freshwater fish Pseudogobio esocinus. We conducted comprehensive phylogeographic analyses based on partial sequences of the mitochondrial DNA cytochrome b (cyt b) gene in 1,212 P. esocinus specimens from 201 locations in 136 river systems along with sequence data from related species. We also examined the entire cyt b sequence and three nuclear gene sequences for a subset of individuals to estimate the phylogenetic relationships and divergence times. Historical processes shaping the present phylogeographic patterns were inferred by the Bayesian skyline plot, mismatch distribution and nested clade phylogeographic analyses. The mitochondrial and nuclear DNA phylogenies showed that the Japanese P. esocinus was not monophyletic and consisted of three largely differentiated groups. Divided by the Central Highlands, two groups were distributed in southwestern Japan, with a third group in northeastern Japan. The estimated divergence time indicated that the framework of the present ranges of the three groups was formed during or before the Early Pliocene. In southwestern Japan, the two groups often occurred sympatrically and included several regional subgroups bounded by major mountain systems uplifted in the early Pleistocene, while in the northeastern Japanese group, recent (since about 0.3 million years ago) range expansion from the northwestern area was inferred. The results suggest the importance of the uplifting of the Central Highlands and interconnection with the continental area in the distribution and divergence of P. esocinus, which may have also influenced regional heterogeneity in the Japanese freshwater fauna. The contrasting phylogeographic patterns between northeastern and southwestern Japan probably reflect their topographic and geohistorical differences, and imply differences in the formation process of freshwater fauna between the two regions.

We describe a new species of spined loach, Cobitis sakahoko, based on the holotype and nine paratypes collected from the Oyodo River system in the southern region of Kyushu Island, Japan. This species is distinguished from its congeners by the following combination of characteristics: lamina circularis at the base of the pectoral fin in adult males, a rectangular plate with a neck in the mid-lower part; the upper segments of the first branched soft ray of the pectoral fin broad; snout relatively short, length 32.1–38.2% of head length; long maxillary barbel, longer than the eye diameter; prepelvic myotome number 14; and line L5 organised in 8–13 oblong or ovoid blotches.

The spawning behaviour of Pseudogobio esocinus was examined in an experimental water tank. Spawning was observed on 66 occasions in five experiments, and one female repeatedly spawned 7–18 times during one night. Four easily distinguishable behavioural phases were recognised during the spawning sequence: phase 1, male nuzzling to the female’s body or face; phase 2, male pursuing a swimming female; phase 3, pair or trio trembling while swimming; and phase 4, spawning (scattering eggs and sperm) near the water surface. The spawning behaviour mainly occurred between pairs (53 times) and occasionally among a trio (13 times). A generalised linear mixed model was used to analyse the relationships between male spawning success and four fixed variables. As a result, more aggressive and larger males tended to be successful spawners. A digital video image of P. esocinus spawning behaviour is available at http://www.momo-p.com/index.php?movieid=momo150129pe01b .

The variability of life history traits in Pseudogobio esocinus was investigated in a single river (Nakagawa River, Kyushu Island, Japan) by comparing two sub-populations inhabiting river reaches with different thermal characteristics. A total of 689 individuals in the upper reach (St. A) and 705 individuals in the lower reach (St. B) were collected from April 2004 to March 2006 at monthly intervals. Water temperatures of both stations were measured by a logger during the period of survey, and these dates clearly showed that St. A was colder than St. B. The sub-population of St. A consisted of six age groups, and the age at maturity was 2 or 3 years. Spawning occurred between April and July, and peaked in May. The sub-population of St. B consisted of four age groups, the age at maturity was 1 or 2 years and spawning occurred between March and July, peaking in April. The length–weight relationship and body condition were also different between the two sub-populations. In conclusion, the sub-population in the colder reach was characterised as longer lived, slower growing and later maturing, with larger body size and higher fecundity compared with the sub-population of the warmer reach. These results showed that the life history traits of P. esocinus were variable under the influence of water temperature, even in a single small river.

We describe the larval and juvenile development of Pike Gudgeon, Pseudogobio esocinus, using laboratory-reared specimens. Newly hatched larvae with a 4.2–4.6 mm body length (BL) and 22–23 + 16 = 38–39 myomeres had melanophores on their head and body. The yolk sac was completely absorbed at 5.0 mm BL. Notochord flexion was initiated at 5.0–6.0 mm BL and completed at 7.3 mm BL. The aggregate number of all fin rays was completed at 11 mm BL. Several rod-like cupulae were observed on the head and lateral side of the body at 4.2–11.0 mm BL and were completely distinguished at 11.0–14.0 mm BL. Squamation was initiated on the caudal body at approximately 11.0 mm BL and completed at 11.0–14.0 mm BL. The newly hatched larvae had well-developed eyes and large pectoral fins.

The genetic population structure of the small cyprinid Hemigrammocypris rasborella, distributed widely in lowlands of western Japan, was examined using partial sequence data of mitochondrial DNA (mtDNA). Molecular phylogenetic analysis revealed that the populations of the western Kyushu region were markedly differentiated from all eastern populations, such that the groups would be comparable to different species; their divergence was inferred to have occurred in the Late Miocene–Pliocene. Also, a largely divergent mtDNA group (with divergence in the early Pleistocene) was found in the Sanyo and northeastern Shikoku regions, forming a secondary contact zone in the western Kinki with the eastern mtDNA group. To date, these aspects of the population structure of H. rasborella appear to be unique among lowland fishes in western Japan. Deeper understanding of the formation processes of freshwater faunas in western Japan will require further comparisons of the phylogeographic patterns and ecological traits of constituent species.

The distributions of seven bitterling species and subspecies—Tanakia lanceolata, T. limbata, Acheilognathus tabira nakamurae, A. rhombeus, Rhodeus ocellatus kurumeus, R. ocellatus ocellatus, and R. atremius atremius—in northern Kyushu were predicted using generalized linear models (GLMs) in order to provide information helpful for conserving native bitterlings and preventing the expansion of alien bitterling species. Predictions were made according to the following procedure: (1) a set of GLMs for each species was formulated using environmental data from 710 sites that were derived using digital maps and GIS software, from which the best fit model for each species was selected using the Akaike information criterion for predicting the fish occurrence, (2) model performance was evaluated based on the receiver-operating characteristics (ROC) analysis using occurrence and environmental data from 362 sites, and (3) potential distributions of the bitterling were analyzed using the best fit models and environmental data for 1,272 sites, of which 200 data points without occurrence data were prepared. The best fit models revealed that 4–6 environmental factors were important in predicting seven bitterling distributions, which was supported by the area under the ROC curve (AUC) values of these fishes ranging from 0.753 to 0.927. The AUC values in model evaluation were significantly greater than 0.5 for six fishes, suggesting the moderate accuracies of these best fit models for predicting the fish distributions. These predictive models can be used for evaluating potential native bitterling richness and the potential distribution expansion of an alien subspecies.

The potential invasiveness of 28 freshwater fishes in northern Kyushu Island, Japan, was evaluated using the Fish Invasiveness Scoring Kit (FISK). The five co-authors scored the level of invasiveness for each species and calculated the total FISK scores; the maximum and minimum scores were then eliminated, and the mean of the remaining three scores was used as the final score for each species. The mean scores ranged from 11.0 (Hypomesus nipponensis) to 31.0 (Cyprinus carpio). The receiver operating characteristic curve indicated that the threshold value between fishes that present a high risk of invasion and the other species were 19.8.

The reproductive ecology of two endangered subspecies of the bitterling Rhodeus atremius was investigated in two rivers in Okayama and Fukuoka Prefectures, Japan. Several other bitterling species, R. ocellatus, Tanakia limbata, T. lanceolata, Acheilognathus tabira and A. rhombeus, are also found in these rivers. Spawning of both R. atremius subspecies was recorded between late March and August 2007, peaking between April and July. Females of R. atremius began to mature at small body size (minimum 25 mm). They had small clutch sizes (maximum 16 eggs) and developed a short ovipositor (mean 13.6 mm), which was used to deposit a bulb-shaped egg of comparatively large size (3.2 mm3, major and minor axes 2.8 and 1.5 mm). Clutch size, ovipositor length and egg size appear to be a function of female body size. Both the body size and ovipositor length of R. atremius with ripe ova and the size and shape of the eggs varied seasonally. Rhodeus a. suigensis achieved maturity at a smaller body size, had a longer ovipositor and produced smaller eggs, with the minor axis being wider relative to the major axis than those of R. a. atremius. Rhodeus atremius may be adapted for depositing small numbers of eggs in a much shallower position inside the gills of smaller host mussels than other bitterling species.