RS JC JapaneseCoral Reef S Galaxea, JCRS, 6: 43 - 46 (2004) ociety

Short communication

Feeding habits of juvenile darkfin hind urodeta () at Iriomote Island, southern Japan

Tadaomi Nakai, Yohei Nakamura, Mitsuhiko Sano and Hisashi Kurokura

Department of Global Agricultural Sciences, Graduate School of Agricultural and Life Sciences, University of Tokyo, Yayoi, Bunkyo, Tokyo 113-8657, Japan Tel: 81-3-5841-8115. Fax: 81-3-5841-5189. Email: [email protected]

Abstract The feeding habits of juvenile serranids, Cephalopholis urodeta, were examined on the basis of 104 specimens ( •…_ 100 mm in total length, TL) collected from coral reef areas at Iriomote Island, Ryukyu Islands, Japan, from May to September between 1998 and 2001. Thirtyone (29.8%) of the specimens collected had empty stomachs, the overall mean stomach fullness index being low (0.94). The most important prey comprised small fishes and shrimps, food preferences not differing among three juvenile size classes ( •…_60, 61-80 and 81-100 mm TL).

Key words Cephalopholis urodeta, Feeding habits, Ontogenetic diet shift, Stomach fullness index

feeding habits of juvenile C. urodeta, including ontogenetic diet shifts with growth. INTRODUCTION The darkfin hind, Cephalopholis urodeta, is a common and widespread species, mainly inhab- iting coral reefs in the west-central Pacific and MATERIALS AND METHODS Indian oceans (Heemstra and Randall 1993). At The study was carried out at Amitori Bay Iriomote Island, Ryukyu Islands, Japan, it is (24° 20'N, 123° 42'E), situated on the western one of the dominant (Epinephelinae) side of Iriomote Island, from May to September species on reef flats and slopes, coexisting with each year between 1998 and 2001 inclusive. A many other , including Epinephelus merra map of the study site was given by Sano (2000). and E. fasciatus (Nakai 2002a). Individuals of C. urodeta less than 100 mm in Although information on the feeding habits of total length (TL) were defined as juvenile by adult C. urodeta is available, the major food examining gonad maturation (Nakai 2002b). items being small fishes and decapod crusta- Juvenile fish were collected from the shallow ceans (Randall and Brock 1960 Hiatt and reef flats and slopes to a depth of 30 m by di- Strasburg 1960; Nakai et al. 2001), the diet of vers using hook and line with fish meal bait and juvenile C. urodeta, as well as juveniles of other scoop nets. Sampling was conducted between serranid species, have received scant attention. 1200 and 1800 h, C. urodeta being a diurnal preda- Studies on the feeding habits of juvenile serranids tor (Nakai 2002b). Immediately after collection, are useful for understanding the overall ecology 10% formalin was injected into the abdominal of groupers and for determining release areas cavity of each specimen, which was then pre- for hatchery-reared serranid juveniles in natu- served on ice. ral environments (Hamasaki et al. 2004). The In the laboratory, within 4 hours of collection, aim of this study, therefore, was to clarify the TL and body weight were measured for each 44 Tadaomi Nakai, Yohei Nakamura, Mitsuhiko Sano and Hisashi Kurokura

specimen to the nearest 1 mm and 0.1 g, re- analysed as fixed factors. Prior to these analy- spectively. Specimens were sorted into 3 size ses, data were transformed to aresine V. classes: ~ 60 mm TL (I), 61-80 mm TL (II), and 81-100 mm TL (III). Food items from the stom- ach contents were sorted and identified to family level under a binocular microscope. The volume RESULTS of each item in the diet was visually estimated for Of the 104 stomachs examined, 73 individu- each specimen to the nearest 0.1 mm3, using a als (70.2%) contained food items, 31 (29.8%) slide with a 1 x 1 mm grid. The relative impor- being empty. The overall mean SFI of speci- tance of each prey item in the diet was ex- mens was 0.94. In terms of mean SFI in each pressed as the mean percentage composition of size class, a relatively low value was obtained in each item by volume (%V), which was calculated sizeclass III (mean ± standard deviation = 0.21 by dividing the sum total of individual volume ± 0.38), compared with those in I (0.95 ± 1.06) percentage of the item by the number of speci- and II (1.44 ± 4.24). However, a significant mens examined (Hobson 1974 Sano et al. 1984). difference was not observed among the three size Specimens with empty stomachs were excluded classes (one-way ANOVA, F2,101=1.69,P=0.19). from this analysis. The overall feeding habits of C. urodeta are To measure the degree of feeding intensity of shown in Table 1. The major food items were each fish, we used the stomach fullness index small fishes and decapod , account- (SFI): ing for 25.1% and 35.8% of the total stomach SFI = [SCW / (BW - SCW)] x 100, contents by volume, respectively. Of fish prey, where SCW is the weight of the stomach con- Gobiidae was the most dominant, although uni- tents and BW is fish body weight. Specimens dentified fishes also accounted for a large por- with empty stomachs were included in the com- tion of the fish dietary component. Among parison of mean SFI among body size classes. decapods, shrimps (predominantly Palaemonidae) A one-way analysis of variance with unequal were the most prevalent prey item (21.3%), fol- replication (ANOVA) (Zar 1999) was used to lowed by hermit crabs (predominantly test for body size effects on SFI and percentage Galatheidae) (10.5%), decapod larvae (4.6%) composition of major food items. Body size was and other crabs (4.0%).

Table 1. Percentage volume (%V) of food items in the diet of juvenile Cephalopholis urodeta (N=73: Total length=49.5-99.8 mm) Feeding habits of juvenile darkfin hind 45

Fig. 1. Percentage volume (%V) of food items in the diets of three size classes of juvenile Cephalopholis urodeta. Size classes I-III defined in text. Number of stomachs examined with food given above each col- umn.

Based on the %V data, differences in the die- urodeta was low (0.94), also similar to that of tary composition of fish and shrimps among adults (0.64; Nakai et al. 2001). size classes were examined (Fig. 1) . A one-way In the present study, juvenile C. urodeta did ANOVA revealed that %V of each food item did not show any clear ontogenetic diet shift with not differ among the three size classes (F2,16 growth, taking mainly fish and shrimps, as also =0 .28, P=0.76 for fish F2,17=0.87, P=0.44 for recorded for adults (Nakai et al. 2001). Similar shrimps). In addition to fish and shrimps, piscivory in juveniles has been reported in larger-sized juveniles (>61 mm TL) fed on Plectropomus leopardus (<100 mm SL) on the Great crabs and hermit crabs, but the proportions of Barrier Reef (St. John 1999) and Myceroperca these food items in the diet were low. microlepis in North Carolina (49-150 mm SL) (Ross and Moser 1995). Juvenile Epinephelus striatus, on the other hand, fed mainly on DISCUSSION gammarideans and hermit crabs (50-109 mm TL) (Heck and Weinstein 1989 Grover et al. About 30% of the C. urodeta specimens exam- 1998), indicating that food habits of juvenile ined had empty stomachs, a finding similar to the serranids is species-specific. situation with juvenile coral trout, Plectropomus leopardus, on the Great Barrier Reef, Australia, 25% with empty stomachs (St. John 1999). Ross and Moser (1995) found that 40% of juvenile ACKNOWLEDGMENTS Mycteropomus leopardus from North Carolina also We are grateful to Hiroyoshi Kohno, Hiroyuki had empty stomachs. These findings indicate Yokochi, Ken Sakihara, Nagahiro Nakazato and that a relatively high proportion of juvenile grou- the Okinawa Regional Research Center, Tokai pers have empty stomachs, as do adults (Brul University, for assisting in field work. Con- e and Rodriguez Canche 1993 Eggleston et al. structive comments on the manuscript from Kou 1998). The overall mean SFI of juvenile C. Ikejima, Masahiro Horinouchi, Graham Hardy, 46 Tadaomi Nakai, Yohei Nakamura, Mitsuhiko Sano and Hisashi Kurokura

and anonymous reviewers were much appreci- southern Japan. Fish Sci 67: 640-643 ated. This study is a contribution from the Nakai T (2002a) Management of fishery resources Okinawa Regional Research Center. for groupers (Serranidae) in Okinawa, southern Japan. Fish Sci 68 suppl: 431-432 Nakai T (2002b) Life history and social system of the darkfin hind Cephalopholis urodeta at REFERENCES Iriomote Island, Ryukyu Islands, Japan. Brule T, Rodriguez Canche LG (1993) Food Ph.D. thesis, University of Tokyo, Tokyo, habits of juvenile red groupers, Epinephelus 137pp (in Japanese) morio (Valenciennes, 1928), from Campeche Randall JE, Brock VE (1960) Observations on Bank, Yukatan, Mexico. Bull Mar Sci 52 : the ecology of epinepheline and lutj anid 772-779 fishes of the Society Islands, with emphasis Eggleston DB, Grover JJ, Lipcius RN (1998) on food habits. Trans Am Fish Soc 89: 9-16 Ontogenetic diet shifts in nassau grouper Ross SW, Moser ML (1995) Life history of juve- trophic linkages and predatory impact. Bull nile gag, Mycteroperca microlepis, in North Mar Sci 63: 111-126 Carolina estuaries. Bull Mar Sci 56: 222- Grover JJ, Eggleston DB, Shenker JM (1998) 237 Transition from pelagic to demersal phase Sano M (2000) Stability of reef fish assem- in early-juvenile nassau grouper, Epinephelus blages~ responses to coral recovery after striatus; pigmentation, squamation, and catastrophic predation by Acanthaster planci. ontogeny of diet. Bull Mar Sci 62 97-113 Mar Ecol Prog Ser 198 121-130 Hamasaki K, Takeuchi H, Shiozawa S, Teruya Sano M, Shimizu M, Nose Y (1984) Food habits K. (2004) Effects of acclimation to the natu- of teleostean reef fishes in Okinawa Island, ral environment before on retention rate, southern Japan. Univ Mus, Univ Tokyo, feeding condition and predation of hatch- Bull 25: 1-128 ery-reared juveniles of the coral trout Zar JH (1999) Biostatistical analysis - 4th ed. Plectropomus leopardus released on a coral Prentice Hall, Upper Saddle River, New reef. Nippon Suisan Gakkaishi 70: 22-30 Jersey. (in Japanese with English abstract) Heck KL Jr., Weinstein MP (1989) Feeding (Received 7 April 2004/Accepted 3 Sept. 2004) habits of juvenile reef fishes associated with Panamanian seagrass meadows. Bull Mar 西表島 におけるニ ジハ タ稚魚の食性 Sci 45~ 629-636 Heemstra PC, Randall JE (1993) FAO Species 中居正 臣 ・中村洋平 ・佐 野光 彦 ・黒倉 寿 Catalogue Vol. 16, Groupers of the World (Family Serranidae, Subfamily Epinephelinae). 東京大学大学 院農学 生命科学研究科 An Annotated and Illustrated Catalogue of the Grouper, Rockcod, Hind, Coral Grouper and 琉球 諸島西表島 にお いて,1998年 か ら2001年 の5- Lyretail Species Known to Date. FAO Fisheries 9月 に採集 した104個 体 を用 いて,ニ ジハ タ稚魚 の食 Synopsis 125, FAO, Rome, 382pp 性 を明 らか に した。31個 体 が 空胃で あ り,採 集 した Hiatt RW, Strasburg DW (1960) Ecological re- 全個体 の胃内容物重量指数は,0.94と 低 い値 を示 した。 lationships of the fish fauna on coral reefs 稚魚 の最 も主要な餌 生物 は魚 類 とエ ビ類であ り,全 胃 of the Marshall Islands. Ecol Monogr 30: 内容 物体積 に占める割 合は46.4%で あった。 さ らに, 65-127 本種 の主要な餌 生物は,成 長 に伴 う明確な変化 を示 さ Hobson ES (1974) Feeding relationships of なかった。 teleostean fishes on coral reefs in Kona, . Fish Bull 72: 915-1031 St. John J (1999) Ontogenetic changes in the diet of the coral reef grouper Plectropomus leopardus (Serranidae) patterns in taxa, size and habitat of prey. Mar Ecol Prog Ser 180: 233-246 Nakai T, Sano M, Kurokura H. (2001) Feeding habits of the darkfin hind Cephalopholis urodeta (Serranidae) at Iriomote Island,