Fisheries Science 63(1), 15-21 (1997) Spawning Cycle of Two Dragonet Species, Calliurichthys japonicus and Repomucenus huguenini, in Tosa Bay, Southern Japan Benjamin Jareta Gonzales,*1,*3 *1Department of Fisheries *2Department of Biology *3State , Faculty of Agriculture, Taniguchi,*1,•õ and Kochi Nankoku, University, Osamu Okamura*2 Kochi , Faculty of Science, Kochi University, 2-5-1 Akebono-cho, College of Palawan -Institute of Marine Sciences , Polytechnic Puerto Nobuhiko Princesa City 5300, 783, Japan Kochi 780, Japan Philippines (Received February 9, 1996) The from japonicus spawn the spawning March was during cmSL bution peak occur the in male. range of at its from Key peak season 34.4-34.6in words: of in Calliurichthys February 1993 spawner, 90-120m peak japonicus in a wild having water reproductive high depth huguenini water depth during in July. Corresponding author. (May) This species matures and reproduces in both sexes. The temperature in July, ranged of R. huguenini 15 spawning, to 30m from 17.4-21.6•Ž water length at the Kochi of females and an salinity temperature early when from ranged at within (October to age with the GI of 12.5cm appears waters in SL its normal to of and distri December with a maturity estimated to C. japonicus samples 34.6-35.0. from investigated Calliurichthys The species to shallower occurs spawning autumn was Prefecture. summer. ground spawner, from 45-120m and in November, huguenini Bay, activity in its spawning Maturity is a biannual spring Repomucenus in Tosa reproductive and extends During 22.4-22.8•Ž the and peak salini depth. maturity, dragonet, Information on the ecological interactions of fish spe cies is a prerequisite for their effective management and conservation. Conservation efforts for fish species should consider not only those species of fishery or of economic importance, but other ecologically important species as well.1) Fish catch compositions during trawling surveys in Tosa Bay reveal a high catch percentage of a diverse array of dragonet fishes. Calliurichthys japonicus and Repomucenus huguenini are the two dragonet species which predominantly occur among these fish catches. Because of their abundance, these two species are likely to play an important ecological role in the benthic communi ty of the Bay. Life history strategies of fishes can be deduced by studying or comparing similar species in the same or simi lar environments.2) It is also important to understand which of the life history characteristics and environmental factors covary. In the present paper, we detail the spawn ing season and length at maturation of C. japonicus and R. huguenini to clarify the dynamics of the benthic com munity of Tosa Bay. In addition, since C. japonicus and R. huguenini are both abundant and are benthic ver tebrates in the Bay, they can also be useful as biological in dicators to detect environmental changes that may occur in the Bay. Therefore, any basic information on the seasonal reproductive timing and length at maturation of both species may assist in detecting or forecasting the vari ous impacts of human-induced changes on the benthic ecosystem of Tosa Bay. •õ and population period Repomucenus 15-30m in November). at 4.5cm SL reproductive ty to is an annual mainly between Bay 17.7 cycle 1992 Tosa Bay Research on dragonets has determined the age and growth of Callionymus lyra;3) the general biology and ecol ogy of Foetorepus altivelis,4) Callionymus belcheri,5) and Diplogrammus xenicus;6) the life-history of Repomucenus beniteguri;7) and the spawning of Repomucenus valencien nei and R. richardsonii8) and Paradiplogrammus enneac tis.9) However, no reports are available on the spawning of two predominant dragonets Calliurichthys japonicus and Repomucenus huguenini occurring in Tosa Bay. Materials and Methods Samples were collected monthly in Tosa Bay from eight depth zones of 15, 30, 45, 60, 90, 120, 150, and 190m dur ing the period from March 1992 to February 1993 (Fig. 1). Sampling of fish specimens was carried out using a beam trawl (7m beam length, 10m total net length with 10mm cod end mesh). The net was towed for about 30 minutes at each depth zone. The gear was operated on board the 20 ton R/V Toyohata Maru of the Usa Marine Biological Institute of Kochi University. The bottom water was sam pled with a Nansen bottle sampler at each station. A revers ing type thermometer mounted on the water sampling bot tle and a conduction type salinometer were used to analyze the proximate environmental parameters. Fish specimens were temporarily stored on ice while on board the vessel and were preserved in 10% formalin solu tion immediately after sorting. In the laboratory, standard length (SL) and body weight (BW) of the fish were meas 16 Gonzales et al. ured to the nearest gonads were The monthly determined, adult and fish months R. sex the male C. organ by of the color) GI was The (g) •~ specimens ?17.0cm Fig. 1. Map of study area and location of trawling survey sites (dotted portion). depth the 2. Monthly collected In dotted changes from C. japonicus, lines). Tosa in gonad Bay, no March female indices (GI) of 1992-February specimens ? 12.1 Calliurichthys 1993 or japonicus (A, were (Fig. female; B, male) (Fig. and To male R. female specimens while depths R. sizes discern has huguenini, in length accurately which different dark Gl=gonad with japonicus, than from and only analysis, 2). confirmed formula: specimens C. length white). japonicus, GI R. genital finally milky SL of male body (elongated the C. for range to such length in elongated was testis and female of collected length 12.1cm used ground distribution samples separately Fig. SL tail fin SL; by longer dorsal Sex the In used and of spawning <9cm relatively using SL3(cm). oocytes initially relatively (rounded were male in of the peak japonicus, first species. length ? spawning the calculated 104/ SL ? 4.5cm the longer both of the determined as the on color ovary of huguenini, the in the (C. and and The a microscope. was relatively males and weight SL) janonicus, shape stage under finrays respectively. during individuals characters filamentous huguenini, of maturation <6cm morphological 0.1g, weighed to the nearest 0.01mg. index (GI) of both species were collected small huguenini, the the observed of and and gonad specimens were The 0.1cm removed mean a wider we analyzed and months 4). and Repomucenus huguenini (C , female; D, male) (Mean•}SE). male specimens ? 17.0cm in SL were collected in March, May, and October (indicated by Spawning in Two Results Spawning Season of C. japonicus The mean monthly GI of female C. japonicus started to increase from January (Fig. 2A). No large individuals (? 12.0cm in SL) were caught in either March or May, but in April, the mean GI value showed a continuation of this trend. A significant increase in female GI was observed it June (Fig. 2A), and the mean monthly GI of female C, japonicus (? 12.0 cm in SL) peaked in July, having high values over the summer, from June to September (Fig. 2A). No large male C. japonicus (? 17.0cm in SL) were col. lected in March or May. The mean monthly GI of male C. japonicus was relatively high in April (Figs. 2B, 3). The male GI increased in June, while a slight decrease was ob served in July (Fig. 2B). The GI of large males (? 17.cm. SL) increased significantly in August reaching the peak it September (Fig. 2B). Mean monthly GI values of both sex. es were low in winter (Figs. 2A, B and 3). A sligt males increase (> smaller-sized SL) male SL) cohorts As larger were increase females (13.0-17.0 These tinued in until and year 3). GI SL. (Fig. value Of was the (5.0-10.8), Samples depths were distributed of throughout the the collected also with with high depths. Large males depths in collected males depth in The males high GIs GIs of July high and collected and almost and a few GIs were at 45-120m, at its from of the collected (Fig. in 45 males However, only in 90m of C. japoni 4D). when peak year large January. (Fig. 60m in deep observed relatively in large and collected also July water July, 45 months were depths September salinity 45-120m In were all also was in were and were highest While speci between however, 45-60m with June September. C. japonicus depth. 45 and (90-120m), in observed appeared in in August were through 17.7 values however, depths the GI in July, ranged 34.4-34.6 at season range 15 to of R. was huguenini in 22.4-22.8•Ž 30 m depths, where and samples of the oocytes observed in July, and R. were characterized (vitellogenic in adult huguenini by a very specimens during narrow of C. May and previtelline stage). Discussion a individuals waters females, 60m was GI female the with water between of these fish samples in August (Fig. 4B). males, temperature samples 17.4-21.6•Ž of reproductive temperature collected. japonicus November space peak the GI of females with Large con their February of that deeper large and June large 4). Some (90-120m) Some large cus in high females er waters (Fig. values zone during were and specimens Bay 45 depths - 60m SL till in GIs 17 the salinity were Most occurred high size the a larger thereafter low GIs. June, and values while smallest mostly year. between (Fig. 4A). The GI in the 120m depth 4C). GI During November, Individual GI relatively remained female 3). highest high fe Species All females were reproductively inactive, having low GI values from March to April, while from May larger ones (approximately 10.0 cm mean SL) with a maximum GI value of 10.2 appeared (Fig. 5). Smaller females (7.0-9.0 cm in SL) began to show higher GIs in July, and these in creased continuously until the peak in November. The most active female spawning cohort in November had a mean SL of 7.0cm, while a larger cohort with a mean SL of 12.0cm showed a decreasing GI in the same month. Testes began to ripen earlier than the ovaries in April (Fig. 5). Comparatively high male GI values occurred from April to May and from September to November. The male GI gradually decreased from December to February of the following year. During the high spawning activity of R. huguenini from October to December, 26% of the total fe male specimens with relatively high GI values (10.3-17.9) had a size range of 4.5-7.2cm SL, while in males, 24% of the total specimens with relatively high GI values (0.3-0.9) had a size range of 4.1 to 7.3cm SL. The smallest length size with a high GI value was 4.5cm SL for both male and female R. huguenini. Only low in (Fig. with The showed 74% were collected from 90m of C. japonicus were collected 120m mens with 12.5cm total all and 3). (8.0-11.9cm sampled the September, decreased female and with individuals to appear (Fig. occurred SL) SL) large males were GI following cm in May cm of both in April and individuals markedly high valuu observed collected (17.0-21.7cm July. GI was (8.5-12.6cm) female marked males in the 14.0cm Dragonet from 34.6-35.0. Spawning Season of R. huguenini The mean monthly GI values of R. huguenini (?4.5cm SL) in both sexes were relatively low from March to Sep tember, but a significant increase occurred from Septem ber to October, reaching a peak in November (Fig. 2C, D). Thereafter, it markedly decreased until February. Spawning of Calliurichthys japonicus Although we lack data for March and May, the GI values of C. japonicus increase from February to June, reaching a peak in July. The mean monthly GI value (Fig. 2A) and the presence of large individuals with high GI values (Fig. 3) suggest that female C. japonicus spawn in the summer, from June to September with the highest reproductive condition being in July. Calliurichthys japonicus may be classified with the group of dragonets that spawn once a year (annual spawner). The peak of its reproductive effort is the same as that of Callionymus enneactis, spawning in July.9) The an nual reproductive frequency of C. japonicus is similar to Callionymus belcheri, spawning once a year. However, C. belcheri spawns between March and May,5) occurring earli er than C. japonicus. These two species were described by Fricke10) as belonging to the same genus-Callionymus. The annual reproductive frequency occurrence of C. japonicus coincides with that of Synchiropus altivelis, which spawns annually between late winter and early spring.4) However, the spawning season of the former differs from that of the latter by occurring in summer. The Synchiropus species groups are easily distinguished from the Callionymus species by their generic differences.11) Furthermore, Fricke10)classified Calliurichthys japonicus, Callionymus enneactis (Paradiplogrammus enneactis in Nakabo,12)) and Callionymus belcheri in one genus-Cal lionymus. In our result, these three species are separated from Synchiropus altivelis by spawning annually in either spring or summer, while the latter species spawns during 18 Fig. 3. Gonzales et al. Relationship between standard length (SL) and gonad index (GI) of Calliurichthys japonicus the colder months of late winter and early spring. Added to this, Synchiropus altivelis is distributed in relatively deeper water (180m) than C. enneactis (0-1m),9,13) C. bel cheri (3-33 m),5) and C. japonicus (45-120). On the other hand, Synchiropus altivelis, Callionymus enneactis, and C. belcheri have a similar geographic distribution (from the waters of Australia to the waters around Japan, 5,12,14)) while Calliurichthys japonicus has a relatively wider ge ographical distribution, occurring from the Western Pacific to Western Australia and the Southeast coast of Africa.10,15)Despite their taxonomic and geographical or bathymetrical distribution differences, these dragonets all in Tosa Bay for March 1992 to February 1993. spawn once a year. Spawning of Repomucenus huguenini The mean monthly GI value shows that R. huguenini spawned from October to December with a peak in Novem ber. The appearance of individuals with high GI values (Fig. 5) indicates that spawning also occurs in May. Repomucenus huguenini is therefore a biannual spawner, which also includes species such as Callionymus sublaevis; spawning in March and October; C. calcaratus, May through June and from October to November;16) and Repomucenus richardsonii, April and November.17) Spawning in Two Dragonet Species 19 Fig. 4. Monthly average standard length (top) and average GI (bottom) of all samples of female (A, B) and male (C, D) Calliurichthys japonicus collected from different zones within its distributional range (45-120m depth). Repomucenus huguenini, besides belonging in the same ge nus with R. richardsonii, also has a similar geographic dis tribution, the former from Hokkaido to the Fast China Sea and the latter from Niigata, Sendai Bay, southward to the South China Sea.14)On the other hand, the congeneric species Callionymus sublaevis and C. calcaratus are also similarly distributed in Australia.16) Despite their taxonom ic and geographical distribution differences, these drago nets all spawn twice a year and have a very similar reproductive timing, occurring either in spring and au tumn, or in summer and autumn. In contrast, though C. enneactis, C. belcheri, C. sublaevis, and C. calcaratus are congeneric species, they exhibit different annual reproduc tive frequencies and seasons. These facts suggest that the seasonal reproductive timing and frequency of breeding per year in dragonets may not be influenced by their taxo nomic relationships and geographical or bathymetrical dis tributions, but may depend on an endogenous reproduc tive cycle, which is under genetic control, synchronized to the environmental conditions of their respective habitats. The oocytes in the vitellogenic stage observed in adult specimens of C. japonicus in July, and R. huguenini dur ing May and November, also suggest that both species were ready to spawn during those periods. In other fish, like Nibea mitsukurii, actual spawning in a tank occurred two months later than the spawning period estimated from GI and microscopic observations of oocytes sampled from the wild.18) This finding is inconsistent with those of the dragonets Callionymus belcheri and Repomucenus huguenini. In C. belcheri, the first discharge of follicles oc curred during the peak of the female GI.5) Repomucenus huguenini spawned spontaneously in a tank from late Sep tember to late November,19) coinciding with the significant increase of male and female GI in the present study. Fur thermore, Takita et al.,9) based on ovarian histological ex aminations, confirmed that the (predicted) spawning period in nature of the dragonet Paradiplogrammus en neactis and that which occurred in the tank were syn chronous. These results support the validity of our esti mates of the spawning seasons of C. japonicus and R. huguenini in the wild, using the GI method. Spawning grounds The depth distribution of C. japonicus has a tendency for larger specimens to occur in deeper waters (Fig. 4A, C). Such phenomena are widespread in the demersal fishes of Tosa Bay.20)Both large male and female C. japonicus were collected from the deeper part of its distributional range, 90-120m, throughout the year. The high GI values (Fig. 4B, D) of large individuals (Fig. 4A, C) between 90-120m depths indicate that this depth range is their main spawn ing ground. The appearance of large males and females with high GIs in 45 and 60m depths in July, during the peak reproductive period, suggest that some of the spawn ing individuals migrate to shallower waters (45-60m) to spawn. Such spawning migration may occur to allow the exploitation of wider habitat ranges. On the other hand, the narrow depth distribution (15-30m) of the R. hugueni 20 Gonzales Standard Fig. S. et al. length (cm) Relationship between standard length (SL) and gonad index (GI) of Repomucenus ni implies that no spawning migration occurs and that the species spawn within its distributional range: in the shal low waters of Tosa Bay. Adult male and female R. huguenini collected from 15m and 30m water depths of Tosa Bay spawned in a tank from late September through late November. 19) The difference in the seasonal reproductive timing and the spawning depth between C. japonicus and R. hugueni ni show that there is no space competition during spawn ing between these two species. huguenini in Tosa Bay for March 1992 to February 1993. Length at Sexual Maturity The relative size at maturation between sexes of C. japonicus was similar to that of other dragonets;4,16,17,21) smaller in females and larger in males. In R. huguenini, the relative size at maturation was similar to that of C. bel cheri 5)being the same in both sexes. Early life-history theo ry suggests that high mortality caused by density-indepen dent effects selects for early reproduction and this strategy would be expected in harsh or unpredictable environ ments.22)The early maturation of R. huguenini, is presuma bly influenced by the environmental changes in the shallow Spawning in Two waters of Tosa Bay and is therefore environmentally less stable. On the contrary, C. japonicus thrives in deeper waters, which are a relatively stable environment, and shows a more delayed maturation. Life history traits are relative to the productivity of their habitats; in high-productivity streams, there tends to be several spawnings in a season, early age at maturity and a short life span. In low-productivity streams, the tendency is for a single spawning in a breeding season, delayed maturity and long life spans.22) These circumstances agree with the results of our study. Repomucenus huguenini, presumably living in a high productivity environment, spawned twice a year, matured early, and has a short life span (about 1 year and 4 months*4). In contrast, C. japoni cas lives in an environment with lower productivity, spawns once a year, and has a delayed maturation and lon ger life span (?3 years; Gonzales, unpublished). Acknowledgments Thanks are due to Associate Professor Kunio Sasaki and his students, and the captain and crew of Toyohata Maru for the field assistance. We also thank Associate Professor Tetsuji Nakabo of Kyoto University for providing us with valuable references. We are grate ful to Dr. Ambok Bolong Abul Munafi for his assistance on fish maturi ty, to the anonymous reviewers and Chris Norman for their helpful com ments on this manuscript. Dragonet 7) 8) 9) 10) 11) 12) 13) 14) 15) 16) References 1) FAO: Report of the expert consultation on utilization and conserva tion of aquatic genetic resources. FAO Fisheries Report No. 491. Food and Agriculture Organization of the United Nations. Rome, Italy, 1993, pp. 3-15. 2) D. A. Roff: Reproductive uncertainty and the evolution of iteropari ty: why don't flatfish put all their eggs in one basket? Can. J. Fish. Aquat. Sci., 38, 968-977 (1981). 3) H. Chang: Age and growth of Callionymus lyra L. J. Mar. Biol. Ass. U.K., 30, 281-296 (1951). 4) M. Akazaki: Biological studies on a dragonet, Synchiropus altivelis 5) 6) (Temminck et Schlegel). Japan. J. Ichthyol., 5, 145-152 (1957) (in Japanese). C. R. Johnson: Biology and ecology of Callionymus belcheri (Pisces: Callionymidae). Copeia, 3, 461-470 (1972). R. Fricke and M. Zaiser: Redescription of Diplogrammus xenicus (Teleostei: Callionymidae) from Miyake-jima, Japan, with ecologi *4 B . J. Gonzales: Ecological studies on the callionymid fishes in Tosa 17) 18) 19) 20) 21) 22) Species 21 cal notes. Japan J. Ichthyol., 29, 253-259 (1982). T. Takai and T. Yoshioka: The life history of the dragonet, Cal lionymus beniteguri in the Seto Inland Sea-I. Egg, larvae and juvenile. J. Shimonoseki Univ. Fisher., 27, 147-154 (1979) (in Japanese). T. Takita and E. Okamoto: Spawning behavior of the two drago nets, Callionymus flagris and C. richardsoni, in an aquarium. Japan J. Ichthyol., 26, 282-288 (1979) (in Japanese). T. Takita, T. Iwamoto, S. Kai, and 1. Sogabe: Maturation and spawning of the dragonet Callionymus enneactis, in an aquarium. Japan J. Ichthyol., 30, 221-226 (1983). R. Fricke: Indo-Pacific Callionymidae. Verlag Von J. Cramer Braunschweig, Germany, 1983, p. 774. R. Fricke: The Kaianus-group of the genus Callionymus (Pisces: Callionymidae) with descriptions of six new species. Proc. Calif. Acad. Sci., 42: 349-377 (1981). T. Nakabo: Fishes of Japan with pictorial keys to the species. Tokai Univ. Press, Tokyo, 1993, p. 1474. B. Gonzales and O. Okamura: Rare occurrence of Bathycalliony mus formosanus (Teleostei: Callionymidae) in Tosa Bay, southern Japan with ecological notes and recent list of callionymids in the Bay. SPCP-Inst. Mar. Sci. Res. J., 2, 28-36 (1995). T. Nakabo: Family Callionymidae, in "The fishes of the Japanese Archipelago" (ed. by H. Masuda, K. Amaoka, C. Araga, T. Uyeno and T. Yoshino), Tokai Univ. Press, Tokyo, 1984, pp. 342-346. J. L. B. Smith: Fishes of the families Draconettidae and Callionymi dae from the Red Sea and Western Indian Ocean. Rhodes Univ., Ichthyol. Bull., 28, 547-564 (1963). C. R. Johnson: The systematics and biology of the Australian species of the genus Callionymus (Pisces: Callionymidae). Ph. D. thesis, University of Queensland, Australia, 1972, p. 217. H. Eda: Life history of callionymid fishes in Omura Bay. Ph. D. thesis, Nagasaki University, Nagasaki, 1994, p. 151 (in Japanese). N. Taniguchi and Y. Okada: Study on the maturation and spawn ing of the nibe-croaker collected from Tosa Bay. Reps. Usa Mar. Biol. Inst., 1, 41-49 (1979). B. Gonzales, O. Okamura, and N. Taniguchi: Spawning behavior of laboratory-reared dragonet, Repomucenus huguenini, and de velopment of its eggs and prolarvae. Suisanzoshoku, 44, 7-15 (1996). H. Horikawa, M. Toriyama, and S. Kishida: Ontogenetic habitat shifts of demersal fishes along depth gradient. Rept. Fish. Res. In vest. Japan. Gov., 27, 61-78 (1991) (in Japanese). T. Takita: Embryonic and larval development of the callionymid fish, Callionymus calliste. Japan. J. Ichthyol., 29, 441-445 (1983). R. J. Wootton: Ecology of teleost fishes. Chapman and Hall, New York, 1990, p. 404. Bay, southern Japan. M. Sc. thesis, Kochi University, Kochi, 1994, p. 80.