Habitat Preference and Feeding Habits of Juvenile Whitespotted Conger Conger Myriaster in the Eastern Seto Inland Sea, Japan

Aquaculture Sci. 58（2），167－179（2010）

Habitat Preference and Feeding Habits of Juvenile Whitespotted Conger Conger myriaster in the Eastern Seto Inland Sea, Japan

1, 2,＊ 2 Shigeaki GORIE and Kazuya NAGASAWA

Abstract: From 2003 to 2008, we tried to collect newly settled juvenile whitespotted conger in four areas of the northeastern Harima Nada Sea, eastern part of the Seto Inland Sea, Japan. We also investigated the relation between the stomach contents of juveniles (78-259 mm TL) and benthos composition at the sampling sites. No juveniles were taken from the sand bottom area, but they were collected from the gravel bottom and sand-mud bottom areas. Main stomach contents of the juveniles were crustaceans, polychaetes and teleosts. Crustaceans were strongly preferred being the most common and numerically important prey. Juveniles first fed on amphipods and, with growth, shifted their prey to decapods, polychaetes and teleosts. CPUEs (catch per unit effort) of the juveniles from the sand-mud bottom were significantly higher than those from the gravel bottom, although the crustacean biomass on the gravel bottom was more abundant than those on the other bottoms. These results suggest that, when the juveniles settle, they prefer the sand-mud bottom to the gravel bottom and choose their habitat not by the food environment but by the sediment type. We consider that the sand-mud bottom area is a nursery ground of the juveniles.

Key words: Conger myriaster; Juvenile; Habitat; Stomach content

The whitespotted conger Conger myriaster is methods for efficient management, it is indispens- one of the commercially important fishes in the able to obtain information on various aspects of the Seto Inland Sea, Japan. Average annual catch life history of the target species. of this fish species from 1985 to 1997 in the sea Whitespotted conger are known to begin off Hyogo Prefecture was about 1700 metric their demersal life before completely finishing tons (mt) but it has decreased to 607 mt in 2007 metamorphosis and then to shrink in length (Anon. 2009). after settlement (Kubota 1961; Asano et al. With implementation of the fishery resource 1978). In the northeastern Harima Nada Sea, management concept, a management approach eastern part of the Seto Inland Sea, this species was started in the mid-1990s for whitespot- occurs as leptocephali from February to May ted conger in the eastern Seto Inland Sea. (Gorie and Tanda 2005). Juveniles of 0+ year Especially, an efficient and stable stock manage- cohort are about 250 mm in total length (TL) in ment is an important subject during the period August, when fish of the 1+ year cohort reach of stock decline. For the efficient utilization of at least 300 mm TL (Gorie and Ohtani 1998). In fish resources, there are various potential man- aquaria, juveniles can be reared by feeding poly- agement plans as reduction of fishing efforts chaetes and fresh fishes (Takai 1959) or even (especially for protection of juveniles), conserva- commercial diets (both moist and dry pellets) tion or creation of nursery grounds, and mass and grow at a high rate (Gorie and Ohtani 1997; release of seedling. In order to choose appropriate Gorie 2008).

Received October 5, 2009: Accepted January 18, 2010. 1 Hyogo Prefectural Technology Center for Agriculture, Forestry, and Fisheries (Fisheries Technology Institute), Hyogo 674-0093, Japan. 2 Graduate School of Biosphere Science, Hiroshima University, Higashi-Hiroshima, Hiroshima 739-8528, Japan. ＊Corresponding author: Tel: (+81) 78-941-8601; Fax: (+81) 78-941-8604; E-mail: [email protected] (S. Gorie). 168 S. Gorie and K. Nagasawa

Only a few studies have been conducted juveniles of whitespotted conger occur in the on the feeding habits of commercial-sized eastern Seto Inland Sea and what environmen- whitespotted conger (Mori and Inoue 1982; tal conditions are related to their habitats. We Matsumiya and Imai 1987; Nabeshima et al. also discuss the feeding habits of the juveniles 1994; Fukuda et al. 1997). They prey mostly and their relation to the fauna of prey. on teleosts, crustaceans and mollusks. Similar results were reported for related species: Materials and Methods European conger Conger conger (Cau and Manconi 1984; Morato et al. 1999; O’Sullivan Fish sampling et al. 2004; Vallisneri et al. 2007) and American The study was conducted in the northeast- conger Conger oceanicus (Levy et al. 1988). ern Harima Nada Sea, eastern part of the Seto However, very little is known about the growth, Inland Sea off Hyogo Prefecture, Japan (Fig. 1). feeding habits and habitat during the juve- The survey region was divided into four areas: nile stage of these species. In Japan, Takai Area A (depth: 13 m), Area B (10 m), Area C (1959) reported that juvenile whitespotted (17 m) and Area D (18 m). conger occurred at 1-17 m in depth on sand- Sampling was conducted to catch juvenile mud bottom to mud bottom areas but detailed whitespotted conger from April to September in data have not been published. He also men- 2003-2008 by a bottom trawl net with a 2.4 mm tioned that the fish of 100-200 mm in TL fed mesh size of a cod end (Gorie and Tanda 2004). on shrimps, fishes, crabs and polychaetes. In Since no fish was caught during daytime and Korea, Huh and Kwak (1998) found that juve- few fish were obtained in civil twilight, sampling niles of the same species feed mainly on fishes, was carried out at night (30 min after sunset shrimps and crabs in a Zostera bed and show and later). The net was towed for 10-20 min at a very constant prey selection. Previously, we a speed of 2 knots using a small boat (9.1 mt). reported that the diet of whitespotted conger Before towing in each survey area, oceano- changed with fish size (Gorie and Tanda 2004). graphic data including water temperatures Nevertheless, information on the habitat and (WT), salinities and depths were recorded with food habit of their juveniles is quite limited, STD (JFE Alec Co., Ltd., Kobe, Japan). and there have been no studies on relations between prey fauna and stomach contents of Bottom sediment sampling and grain size analysis juveniles. Bottom sediments were sampled with a SK In this paper, we report where newly settled Type bottom sampler (Rigo Co., Ltd., Tokyo,

134°50´E 135°00´E

10m Kakogawa B Sea of Japan C 20m Kami Is. 10 km 34°40´N A D Akashi

No.10 buoy Pacific Ocean Harima Nada Sea

Fig. 1. Location of the survey areas in the eastern Seto Inland Sea, Japan. Habitat and Feeding Habits of Conger myriaster 169

Japan) in Areas B-D on 3 July 2003 and in Area Stomach content analysis A on 22 August 2003. Grain sizes were analyzed In May to August 2004, juveniles soon after based on the Japanese Geotechnical Society being caught were fixed in 5-10% buffered sea- Standard (JSF T 131-1990) and sorted by the water formalin. For stomach content analysis, Wentworth grain size scale. Losses on ignition 30, 106 and 120 fish were selected randomly were measured after the sediments were heated from Areas A-C, respectively (Table 1). The fish at 600℃ for 2 hours. were measured for total length (TL) and body weight (BW) and the stomachs were removed Benthos sampling and their composition and examined for contents. Prey items from Benthos were sampled before towing a trawl each stomach were identified to one of four net in 2004 with a Smith-MacIntyre bottom sam- groups (polychaetes, crustaceans, teleosts pler (22×22 cm, Rigo Co., Ltd., Tokyo, Japan) and others) and expressed as numerical and and by a sledge net (Fig. 2) for 1 min towing at weight percent by sampling date and fish size. a speed of 2 knots. The samples caught using Moreover, crustaceans were classified into the bottom sampler were sieved with a 1 mm amphipods, decapods or others. aperture size and stored in 5-10% buffered sea- Ivlev’s selectivity index (Ei) was calculated -1 water formalin. The benthos composition in as follows: Ei = (ri-pi)・(ri+pi) , where ri is the these samples is expressed as numerical and percentage of prey item i in the fish stomachs, weight percent for five categories: polychaetes, and pi is the percentage of item i in the benthos gastropods, bivalves, crustaceans and cephalo- samples collected with the Smith-MacIntyre chordates. Since these animals accounted for bottom sampler. The index can take values more than 95% of total individuals, other organ- between -1 and 1. A value between 0 and 1 isms were excluded from the analysis. The means that a fish has a positive preference for benthos sampled with the sledge net were cat- the prey item i, while a value between 0 and -1 egorized into one of eight groups: nematodes, shows a negative preference. If a value is 0, a polychaetes, gastropods, bivalves, crustaceans, fish has no preference to the item i. chaetognaths, ophiuroids and teleosts. Statistical analysis Differences in mean CPUE (catch per unit

Outer net, 5mm effort) and in prey density were explored using the Steel-Dwass test between the three survey Inner net, GG50: 0.35mm areas where juveniles were collected.

h=0.4m, w=0.6m Warp 2m Results

Sled

0.7m Chain (6mm) Fish distribution and oceanographic features Juvenile whitespotted conger were collected Fig. 2. The sledge net used to collect epifaunal benthos in in Areas A-C, but no fish was taken in Area the eastern Seto Inland Sea. D. The juveniles were collected from May to

Table 1. Number and size of fish examined for stomach contents Date Area A Area B Area C in 2004 N caught TL (SD, range) N examined N caught TL (SD, range) N examined N caught TL (SD, range) N examined 12, 14 May 7 86 ( 3.5, 81- 91) 1 97 85 ( 3.7, 76- 95) 10 18 84 ( 3.4, 77- 91) 5 23, 24 June 91 120 (13.4, 94-152) 20 181 112 (13.6, 86-151) 19 146 104 (10.5, 85-147) 40 21, 22 July 9 152 (22.8, 128-188) 4 374 139 (19.7, 84-195) 38 154 135 (21.5, 100-195) 40 24, 25 Aug. 15 164 (18.6, 133-208) 5 231 169 (20.7, 105-262) 39 142 167 (23.7, 124-258) 35 Total 122 30 883 106 460 120 N, sample size; TL, mean total length (mm); SD, standard deviation (mm). 170 S. Gorie and K. Nagasawa

Table 2. Mean CPUE, WT and salinity of the fishing surveys Area A Area B Area C Area D Date CPUE WT (℃) salinity CPUE WT (℃) salinity CPUE WT (℃) salinity CPUE WT (℃) salinity 2003/4/9 0 0 10.3 32.9 2003/4/22 0 11.7 32.8 0 2003/5/16 2 15.6 32.2 9 2003/6/13 11 19.5 32.0 8 18.3 32.3 0 19.3 32.4 2003/7/10 94 21.8 31.5 52 21.3 31.9 0 22.3 31.9 2003/8/25 57 25.5 31.0 47 25.0 31.3 0 25.7 31.3 2004/5/12, 14 4 16.7 32.2 97 15.6 32.2 18 15.2 32.4 2004/6/23, 24 44 20.6 31.8 181 20.8 31.6 146 20.0 31.9 2004/7/21, 22 5 24.3 31.7 374 24.2 31.6 154 23.8 31.7 2004/8/24, 25 8 26.5 32.0 231 27.2 31.5 142 26.9 31.7 2005/5/30, 31 3 17.2 31.8 11 16.6 31.7 58 16.2 31.7 0 16.7 31.8 2005/8/24 7 25.3 32.2 82 25.7 32.1 294 25.0 32.1 2005/9/21, 22 12 26.0 32.1 342 26.1 32.1 180 25.9 32.2 0 2006/5/18 0 13.9 32.1 0 13.9 32.2 2006/6/5 8 17.7 31.8 16 16.9 31.7 38 16.2 31.9 2006/7/19, 20 362 23.0 30.4 103 22.0 31.2 129 21.3 31.5 2006/9/14, 15 29 26.2 31.5 212 26.6 30.9 102 26.5 31.0 2007/5/23, 24 0 16.7 32.4 25 15.8 32.5 31 15.8 32.5 2007/7/ 25, 26 9 23.1 31.5 262 23.2 31.7 89 23.0 32.1 2007/9/27 20 27.3 31.9 210 27.4 31.7 40 27.3 31.6 2008/5/26, 27 9 16.4 32.2 78 16.4 32.2 23 15.8 32.3 2008/6/25 36 17.7 31.8 58 20.0 31.7 40 19.3 31.9 2008/9/24, 25 16 25.5 32.6 242 26.1 32.4 42 26.0 32.5 CPUE, mean number of fish caught per 20 min towing; blank, no data. WT and salinity profiles express the data of the sea bottom. All of the surveys were conducted at night (30 min after sunset or later).

September but not in April (Table 2). Mean (%) CPUEs in Areas B and C were significantly 100 higher than that in Area A (Steel-Dwass test, P 0.003), but no significant difference was 80 observed between Areas B and C. The juveniles sampled ranged from 76-95 mm TL in 60 May, 85-152 mm in June, 84-195 mm in July, and 105-262 mm in August. Small-sized juveniles 40 (76-84 mm TL) were sampled over a protracted period from May to July. 20 The bottom WTs and salinities where the 0 juveniles were caught ranged from 15.2-27.3 ℃ Area A B C D and 30.4-32.5, respectively. Fig. 3. Grain size compositions of the survey areas in the eastern Seto Inland Sea. , Gravel and pebble (2 mm<); Bottom sediment type , Coarse sand (0.5-2 mm); , Medium sand The grain size composition of the bottom (0.25-0.5 mm); , Fine sand (0.063-0.25 mm); , Silt was different among the survey areas (Fig. and clay (<0.063 mm). 3). Gravel and pebbles were the major components in Area A but their proportions were were found in Areas B and C but they were not low in Areas B and C, where medium-coarse observed in Area D. Both median and maxi- sand predominated. A small amount of gravel mum size of particles were the largest in diam- and pebbles were also found in Area D, where eter in Area A and the smallest in Area D (Table medium sand was abundant. Some silt and clay 3). Losses on ignition ranged from 2.0-2.9% in Habitat and Feeding Habits of Conger myriaster 171

Table 3. Bottom sediment of the survey areas the four survey areas. Although the depths Area A B C D were different, the bottom of Areas B and C was Median particle diameter (mm) 2.39 0.47 0.41 0.38 similar, and the bottom types of each survey Max diameter (mm) 26.5 19.0 19.0 9.5 area were classified as follows: gravel bottom in Loss on Ignition (%) 2.9 2.3 2.3 2.0 Area A, sand-mud bottom in Areas B and C, and sand bottom in Area D. (%) 100

0 14 May 24 June 22 July 25 Aug. 14 May 24 June 22 July 25 Aug. (%) Area A 100

0 12 May 23 June 21 July 24 Aug. 12 May 23 June 21 July 24 Aug. (%) Area B 100

0 12 May 23 June 21 July 24 Aug.12 May 23 June 21 July 24 Aug. Area C Number percent Weight percent

Fig. 4. Compositions of various groups of benthos collected with the Smith-MacIntyre bottom sampler. Left, number percent; right, weight percent; , Polychaetes; , Gastropods; , Bivalves; , Crustaceans; , Cephalochordates. 172 S. Gorie and K. Nagasawa

Benthos composition Stomach contents There were some differences in composition A total of 256 juveniles were examined for of the benthos collected by the Smith-MacIntyre their stomach contents (Table 1). Throughout bottom sampler between Areas A-C where juve- the survey period, crustaceans were very nile whitespotted conger were collected (Fig. 4). common by number in Areas A-C (Fig. 6). Both In Area A, crustaceans were the most numerous, polychaetes and fishes were also important by but cephalochordates (lancelet Branchiostoma weight in Areas B and C, where polychaetes belcheri) and bivalves were major components showed the highest percentage in June. by weight. In Areas B and C, polychaetes pre- The stomach contents varied between dominated by both number and weight. The the survey areas and with fish size (Fig. 7). biomasses of polychaetes were similar in these Crustaceans were highly predominant in all three areas. The biomass of crustaceans in Area fish sizes in Area A. They were also abundantly A was more abundant than those in the other preyed on in Areas B and C, but the importance areas but no significant difference was observed of polychaetes and teleosts increased by weight (Steel-Dwass test, P 0.05, Table 4). in these areas. Polychaetes were found in the A wide variety of benthos was collected by fish of 80-210 mm TL but were not observed the sledge net (Fig. 5). In Area A, crustaceans in the stomach contents of fish over 220 mm were dominant by number but bivalves and tele- TL. Teleosts as prey were detected in the fish osts were also important by weight. In Areas B over 90 mm TL. Of the crustaceans found, both and C, the proportions of four groups (bivalves, amphipods and decapods were very dominant, crustaceans, polychaetes and nematodes) were but their importance was different by size: the high by number, whereas those of crustaceans proportion of amphipods and decapods declined and ophiuroids were high by weight in these and increased, respectively, with an increase areas, respectively. The composition consider- in fish size (Fig. 8). This trend was found in all ably fluctuated by sampling date. survey areas. Based on the bottom sediment type and the The main amphipods preyed on were podo- benthos composition, Area A is characterized cerids, aorids and pontogeneiids by number. by a gravel bottom with abundant crustaceans Oedicerotids were also detected by weight. and bivalves, and both Areas B and C by sand- As for decapods, penaeids, crangonids and mud bottom with high abundance of various processids mainly occurred by number but benthos, such as polychaetes, bivalves, crusta- pasiphaeids were important by weight. The ceans and ophiuroids. Areas B and C are con- teleosts eaten were gobiids and callionymids sidered to have a similar benthos composition by number, but engraulids were also important and bottom sediment type. by weight. Polychaetes that were preyed on consisted of errantians and sedentarians, and Table 4. Benthos density of the survey areas the errantians predominated (83% by number). Polychaetes Crustaceans Date Sand grains were rarely observed. in 2004 N/m2 W/m2 N/m2 W/m2 Based on the Ivlev’s selectivity index, the 14 May 3,079 16,419 35,310 34,244 24 June 3,574 16,959 10,496 8,281 juveniles showed a strong preference for crus- Area A 22 July 1,033 3,963 9,897 2,163 taceans (Table 5). The crustaceans showed a 25 Aug. 579 5,901 579 2,754 high value throughout the period, regardless 12 May 826 3,112 413 138 of the survey areas. The percentage of empty 23 June 4,566 6,182 124 23 Area B 21 July 4,773 3,886 165 43 stomachs ranged from 10-72% in small-sized fish 24 Aug. 2,748 44,727 165 1,128 (less than 100 mm TL), but few empty stom- 12 May 682 839 248 184 achs were found in the fish larger than 100 mm 23 June 723 4,585 289 314 Area C TL. No particular trend in percentage of empty 21 July 1,281 919 227 68 24 Aug. 1,426 2,713 124 450 stomachs was observed among the three N, number; W, weight (mg). survey areas. Habitat and Feeding Habits of Conger myriaster 173

(%) 100

0 14 May 24 June 22 July 25 Aug. 14 May 24 June 22 July 25 Aug. (%) Area A 100

0 12 May 23 June 21 July 24 Aug. 12 May 23 June 21 July 24 Aug. (%) Area B 100

0 12 May 23 June 21 July 24 Aug. 12 May 23 June 21 July 24 Aug. Area C Number percent Weight percent

Fig. 5. Compositions of various groups of benthos collected with the sledge net. Left, number percent; right, weight percent; , Nematodes; , Polychaetes; , Gastropods; , Bivalves; , Crustaceans; , Chaetognaths; , Ophiuroids; , Teleosts.

To summarize these results, juvenile whites- on amphipods after starting their demersal life potted conger feed mainly on crustaceans, and shift their prey, with growth, from amphi- polychaetes and teleosts and show a strong pods to decapods, polychaetes and teleosts. preference for crustaceans. The juveniles prey 174 S. Gorie and K. Nagasawa

(%) 1204 5 1204 5 100

0 14 May 24 June 22 July 25 Aug. 14 May 24 June 22 July 25 Aug. Area A (%) 10 19 38 39 10 19 38 39 100

0 12 May 23 June 21 July 24 Aug. 12 May 23 June 21 July 24 Aug. Area B

(%) 5404035 5404035 100

0 12 May 23 June 21 July 24 Aug. 12 May 23 June 21 July 24 Aug. Area C Number percent Weight percent Fig. 6. Compositions of stomach contents of juvenile whitespotted conger in the eastern Seto Inland Sea. Left, number percent; right, weight percent; , Polychaetes; , Crustaceans; , Teleosts; , Others. Sample size is given for each sampling date.

that newly settled juvenile whitespotted conger Discussion occur at 10-20 m in depth on gravel bottom (Area A) and sand-mud bottom areas (Areas B and C). The present study has shown for the first time They were not sampled from the sand bottom Habitat and Feeding Habits of Conger myriaster 175

(%) 1246521214 2 1246254 2121 100

0 Area A (%) 1 878888888888 71 0101 1 8 78888888888 71 0101 100

0 Area B

(%) 1 7 14 18 11 10 10 11 127531 7 001 1 1 1 7 14 18 11 10 10 11 12 7 7 5 3 1 0 1 1 0 1 100

0 70- 90- 110- 130- 150- 170- 190- 210- 230- 250- 70- 90- 110- 130- 150- 170- 190- 210- 230- 250- 80- 100- 120- 140- 160- 180- 200- 220- 240- 80- 100- 120- 140- 160- 180- 200- 220- 240- Area C Number percent Weight percent

Fig. 7. Compositions of stomach contents of juvenile whitespotted conger of different total length classes. Left, number percent; right, weight percent; , Polychaetes; , Crustaceans; , Teleosts; , Others. Sample size is given for each size class.

(Area D). The juveniles sampled in May and This study also suggests that juvenile whites- from May to July ranged from 76-95 mm and 76- potted conger exhibit a higher preference for 84 mm TL, respectively, suggesting a protracted sand-mud bottoms than gravel bottoms, because period of settlement after metamorphosis. their mean CPUEs from the sand-mud bottom 176 S. Gorie and K. Nagasawa

(%) 1246521214 2 12462521214 100

0 Area A (%) 187870101878888888888 71 0101 1 888888888 1 100

0 Area B

(%) 1 7 14 18 11 10 10 11 127531 7 01 1 0 1 1 7 14 18 11 10 10 11 127531 7 001 1 1 100

Fig. 8. Compositions of amphipods, decapods and other crustaceans in the stomach contents of juvenile whitespotted conger of different total length classes. Left, number percent; right, weight percent; , Amphipods; , Decapods; , Other crustaceans. Sample size is given for each size class.

(Areas B and C) were significantly higher than areas were slightly different (10 and 17 m that from the gravel bottom (Area A). There respectively), but both the environmental condi- was no significant difference in mean CPUE tions and the benthic fauna were very similar between Areas B and C. Water depths in these to each other. The observed difference in water Habitat and Feeding Habits of Conger myriaster 177

Table 5. Ivlev’s selectivity index showed a strong preference for crustaceans Date Polychaetes Crustaceans (Table 5) and preyed mainly on crustaceans, N in 2004 Number Weight Number Weight polychaetes and teleosts, indicating that juveniles ri 0.0 0.0 100.0 97.1 show selective feeding habits for prey animals 14 May pi 7.3 5.2 83.7 10.8 1 (Fig. 6). Crustaceans are numerically impor- E -1.00 -1.00 0.09 0.80 i tant, while polychaetes and teleosts are quan- r 1.0 2.0 98.3 77.9 i titatively significant. The juveniles also shifted 24 June pi 23.1 12.1 67.8 5.9 20 Ei -0.92 -0.72 0.18 0.86 their food items, with growth, from amphipods Area A ri 0.0 0.6 98.7 52.1 to decapods, polychaetes and teleosts (Figs. 7, 22 July pi 8.5 23.7 81.0 13.0 4 8). Crustaceans were heavily preyed on in Area Ei -1.00 -0.95 0.10 0.60 A, whereas polychaetes were fed on in Areas B ri 0.0 0.0 87.5 87.5 and C (Figs. 6, 7). In these areas, crustaceans 25 Aug. p 28.3 5.3 28.3 2.5 5 i and/or polychaetes were most abundant and the Ei -1.00 -1.00 0.51 0.94 stomach contents of the juveniles correlated with ri 0.0 47.0 100.0 47.7 12 May pi 54.8 65.3 27.4 2.9 10 the available crustacean-polychaetes composition Ei -1.00 -0.16 0.57 0.89 (Figs. 4, 6). This result suggests that the juve- ri 8.3 52.9 63.9 13.5 niles opportunistically feed on crustaceans and 23 June pi 87.0 14.2 2.4 0.1 19 polychaetes in the areas where these animals are Ei -0.83 0.58 0.93 0.99 Area B more abundant than other prey animals. r 7.2 32.6 68.1 10.3 i Although nothing has been reported on 21 July pi 49.1 35.0 1.7 0.4 38 Ei -0.74 -0.04 0.95 0.93 the specific feeding behavior of whitespotted

ri 4.1 21.1 80.8 63.7 conger, epifaunal polychaetes were commonly 24 Aug. pi 76.9 82.3 4.6 2.1 39 found in the juvenile’s stomachs and sand grains Ei -0.90 -0.59 0.89 0.94 were rarely observed. This implies that the juve- ri 0.0 42.3 92.9 34.9 niles have a foraging behavior for epifaunal ani- 12 May p 30.0 40.0 10.9 8.8 5 i mals rather than buried infaunal animals. Ei -1.00 0.03 0.79 0.60 The CPUEs of juvenile whitespotted conger ri 10.7 52.6 85.7 43.2 23 June pi 61.4 47.0 24.6 3.2 40 were higher in Areas B and C than in Area A Ei -0.70 0.06 0.55 0.86 (Table 2), although crustaceans predominated Area C ri 4.3 21.7 83.9 41.1 in Area A and their biomass was more abun- 21 July pi 77.5 41.3 13.8 3.1 40 dant than in other areas (Fig. 4; Table 4). These E -0.89 -0.31 0.72 0.86 i results suggest that the juveniles choose their r 1.9 12.2 91.6 77.1 i habitats not by the food environment but by the 24 Aug. pi 86.3 84.2 7.5 14.0 35 Ei -0.96 -0.75 0.85 0.69 bottom sediment type. ri, percent of prey item i from the fish stomachs; pi, percent We previously reported the food habits of of item i in the benthos samples with the Smith-MacIntyre juvenile whitespotted conger from the north- bottom sampler; Ei, Ivlev’s selectivity index; N, Number of fish examined. eastern Harima Nada Sea (Gorie and Tanda 2004). Huh and Kwak (1998) also reported depth may give no impact on the distribution the stomach contents of juveniles of the same and abundance of the juveniles in these areas. species caught in an eelgrass (Zostera) bed in As for the benthic fauna of the survey areas, Korean waters. The results of these studies polychaetes, crustaceans and mollusks (gas- are slightly different from those in the pres- tropods and bivalves) were most commonly ent study. A higher variation in percentage of sampled (Figs. 4, 5). The biomass of polychaetes prey items was observed in the Harima Nada was similar in Areas A-C and that of crustaceans Sea (Gorie and Tanda 2004), and the juveniles was abundant in Area A (Table 4). The stom- in Korean waters consumed fishes, crusta- ach contents of the juveniles, however, did not ceans and gastropods but did not feed on poly- accord with the benthic fauna: the juveniles chaetes. These differences may be due to the 178 S. Gorie and K. Nagasawa environment of sampling locations between Gorie, S. and T. Ohtani (1997) Growth of the juvenile Japanese (sand-mud bottom) and Korean white-spotted congers under experimental conditions. Suisanzoshoku, 45, 485-488 (in Japanese with English (Zostera bed) waters and/or the smaller sample abstract). size [N=1-18 (mean=6) for each 10 mm size class] Gorie, S. and T. Ohtani (1998) Seasonal changes in the in this study than in the previous one [N=1-26 length compositions of the white-spotted conger, (mean=12)] (Gorie and Tanda 2004). Conger myriaster, caught by the small trawl fishery in Harima-Nada and the possible use of enlarged mesh In conclusion, newly settled juvenile whites- size for resource management. Suisanzoshoku, 46, potted conger were collected from the gravel 495-501 (in Japanese with English abstract). bottom and the sand-mud bottom areas, where Gorie, S. and M. Tanda (2004) Growth and stomach con- crustaceans and bivalves commonly occurred tents of juvenile white-spotted conger Conger myriaster. Suisanzoshoku, 52, 139-144. and various types of benthos were abundant, Gorie, S. and M. Tanda (2005) Occurrence of leptocepha- respectively. The sand-mud bottom area was pre- lus larvae of white-spotted conger Conger myriaster in ferred by the juveniles and is considered to be an northeast Harima Nada, Seto Inland Sea. Bull. Hyogo important nursery ground. The juveniles showed Pref. Tech. Cent. Agr. Forest. Fish. (Fish. Sec.), 38, 1-5 (in Japanese with English abstract). a strong preference for crustaceans and selective Gorie, S. (2008) Juvenile white-spotted conger Conger myr- feeding habits for crustaceans, polychaetes and iaster can be fed on commercial extrude pellet (EP). teleosts. The juveniles potentially choose their Bull. Hyogo Pref. Tech. Cent. Agr. Forest. Fish. (Fish. habitats by the bottom sediment type rather than Sec.), 40, 101-104 (in Japanese with English abstract). Huh, S. H. and S. N. Kwak (1998) Feeding habits of Conger the food environment. Further clarification of myriaster in the eelgrass (Zostera marina) bed in the habitat requirements and feeding habits is Kwangyang Bay. J. Korean Fish. Soc., 31, 665-672 (in needed at the different depths and under various Korean with English abstract). environmental conditions in the Seto Inland Sea Kubota, S. (1961) Studies on the ecology, growth and metamorphosis in conger eel, Conger myriaster (Brevoort). to identify the key habitats used by this species. J. Fac. Fish., Pref. Univ. Mie, 5, 190-370 (in Japanese). Levy, A., K. W. Able, C. B. Grimes and P. Hood (1988) Acknowledgments Biology of the conger eel Conger oceanicus in the Mid- Atlantic Bight II. Foods and feeding ecology. Mar. Biol., 98, 597-600. We thank to Mr. Sadami Kitano and Ms. Matsumiya, Y. and M. Imai (1987) Ecology and abundance Minako Hanaoka, Marine Ecological Institute of conger eel Conger myriaster in Shijiki Bay, Hirado Co., Ltd. for their assistance with identification Island. Nippon Suisan Gakkaishi, 53, 2127-2131. of the benthos and stomach contents. Thanks Morato, T., E. Solà, M. P. Grós and G. Menezes (1999) Diets of forkbeard (Phycis phycis) and conger eel (Conger also go to Mr. Haruyoshi Nakasuji, the captain conger) off the Azores during spring of 1996 and 1997. of research vessel Chidori, and the crews for Arquipélago, Life and Marine Sciences, 17A, 51-64. their help during the survey. Mori, K. and S. Inoue (1982) Ecological notes on the common Japanese conger Conger myriaster (Brevoort). Bull. Fac. Fish., Mie Univ., 9, 179-189 (in Japanese References with English abstract). Nabeshima, Y., T. Abe and T. Kusakabe (1994) Food envi- Anonymous (2009) Annual Statistical Report on Agriculture, ronment in Osaka Bay and feeding habit of whitespot- Forestry and Fisheries of Hyogo Prefecture, Hyogo Assoc. ted conger. Rep. 26th Meet. Inland Sea Fish. Res. Agric. Forest. Statistics, Kobe, pp. 166-201 (in Japanese). Nansei Block., 78-89 (in Japanese). Asano, H., Y. Kubo and S. Yoshimatsu (1978) On the mor- O’Sullivan, S., C. Moriarty and J. Davenport (2004) Analysis phological change and the behavior of the leptoceph- of the stomach contents of the European conger eel ali of Conger myriaster during the period of rearing Conger conger in Irish waters. J. Mar. Biol. Ass. U.K., experiment. Mem. Fac. Agr., Kinki Univ., 11, 25-31 84, 823-826. (in Japanese with English abstract). Takai, T. (1959) Studies on the morphology, ecology and Cau, A. and P. Manconi (1984) Relationship of feeding, culture of the important apodal fishes, Muraenesox reproductive cycle and bathymetric distribution in cinereus (Forskål) and Conger myriaster (Brevoort). J. Conger conger. Mar. Biol., 81, 147-151. Shimonoseki College Fish., 8, 209-555 (in Japanese). Fukuda, T., N. Masunari and M. Hamazaki (1997) Eating Vallisneri, M., M. Scapolatempo and C. Piccinetti (2007) habits of the conger eel in the eastern waters of Preliminary biological data on the northeast Okayama Prefecture. Bull. Fish. Exp. Stn. Okayama Mediterranean conger eel Conger conger L., 1758. Bol. Pref., 12, 1-8 (in Japanese with English abstract). Inst. Esp. Oceanogr., 23, 111-114. Habitat and Feeding Habits of Conger myriaster 179

瀬戸内海東部海域におけるマアナゴ稚魚の生息域と食性

五利江重昭・長澤和也

マアナゴ稚魚が生息する海域の底質・餌料環境を明らかにするため，2003－2008年に瀬戸内海の播磨灘北東部海域において，稚魚の出現状況と底質の粒度組成，ベントスと胃内容物との関連を調べた。稚魚が採集された海域は，砂泥底と砂礫底であった。主な胃内容物は甲殻類，多毛類，魚類であり，甲殻類は個体数からみた重要餌生物，多毛類と魚類は湿重量からみた重要餌生物であった。マアナゴは甲殻類に対する強い嗜好性を持っているが，着底後まずヨコエビ類を捕食し，成長とともにエビ・カニ類や多毛類，魚類へと移行していった。稚魚の生息域には，これらの他にさまざまな動物群が生息しているが，胃内容物には一貫して甲殻類，多毛類，魚類が出現するため，これらを選択的に，またこれらのうち生息域に多数出現する動物群を日和見的に摂餌していると思われた。またマアナゴはその生息域を，餌料環境よりもむしろ底質環境で選択していると考えられた。