Reproductive strategy and recruitment dynamics of
amphidromous goby Sicyopterus japonicus as revealed
by otolith microstructure
K. N. SHEN* AND W. N. TZENG*†‡*Institute of Fisheries Science, College of Life Science, National Taiwan University, Taipei 106, Taiwan and†Department of Life Science, College of Life Science, National
Taiwan University, Taipei 106, Taiwan
(Received 21 December 2007, Accepted 19 September 2008)
The daily ages of 312 of 879 newly recruited postlarvae of Sicyopterus japonicus, collected from the Shuang-Chi Estuary in north-eastern Taiwan during February 1996 to April 1997, were determined from daily growth increments in their otoliths. Pelagic larval duration, growth rate of the marine larval stage and hatching dates were estimated, and recruitment timing was linked to environmental factors. The meanS.D. total length (LT) and daily ages of S. japonicus at recruitment to the estuary were estimated to be 3395 131 mm (range 307 to 381, n ¼ 317) and 16372 1279 days (range 130 to 198, n ¼ 312), respectively. The recruitment of S. japonicuslarvae is size dependent not age dependent because LTof the larvae is independent of age at recruitment. Periodic analysis indicated that LT and growth rate of the larvae were inversely correlated with the age at recruitment, which means that the fast-growing individuals recruited earlier. The growth rate of S. japonicus in the marine larval stage was synchronous with marine productivity in this subtropical area, i.e. the spring cohort recruited in the autumn had a higher growth rate than the autumn cohort recruited the following spring. The main spawning season of S. japonicus as backcalculated from otolith daily increments was in autumn, a relatively low productivity period compared with spring. During this season, there were fewer competitors and predators than in the more productive spring. The recruitment of 95% of postlarva coincided with low salinity (14) and low water temperature (23° C) in the river mouth that provided a buffer area for the adaptation of the larvae for upstream migration. This unique reproduction strategy and prolonged larval duration facilitated the widespread distribution of
the fish along the coasts of East Asia. #2008 The Authors
Journal compilation#2008 The Fisheries Society of the British Isles Key words: daily growth increment; goby; pelagic larval duration.
INTRODUCTION
Sicyopterus japonicus (Tanaka) is an amphidromous goby, mainly distributed in
Japan and Taiwan (Shen et al., 1998; Shen & Tzeng, 2002; Watanabe et al., 2006). Amphidromous fishes migrate between marine and fresh water at some stage in their life history, but the migration is not for spawning (Myers, 1949). They spawn in streams, and their newly hatched larvae drift with the downstream
‡Author to whom correspondence should be addressed. Tel.:þ886 2 33662887; fax: þ886 2 23639570; email: wnt@ccms.ntu.edu.tw
doi:10.1111/j.1095-8649.2008.02102.x, available online at http://www.blackwell-synergy.com
current to the marine environment in their early life history and return to the stream at the postlarval stage (Shen & Tzeng, 2002). Every year, >5–10 million transparent postlarvae of S. japonicus were estimated to recruit from the ocean to streams in eastern Taiwan. An estimated 1 366 752 recruiting individuals of
S. japonicus were recorded during the main fishing season (30 January to 7
May) in 1997 in Hsuikuluan River in mid-eastern Taiwan (Shiao, 1998). The recruitment of amphidromous gobies is known to be correlated with the lunar phase (Delacroix & Champeau, 1992; Shiao, 1998; Keith, 2003; Hoareau et al., 2007a). For example, S. japonicus is recruited mainly in the evening between the last quarter and the first quarter of the lunar phase during spring (Shiao, 1998). After developing pigment and metamorphosing to a benthic life in the lower reaches of the river for a few days, they start their upstream migration (Shen & Tzeng, 2002), climbing even the walls of dams. At this stage, however, they are preyed upon by humans, birds (e.g. striated heron Butorides striatus L. and little egret Egretta garzetta L.) and other carnivorous fishes, such as Eleotris
acanthopomaBleeker and freshwater eels (Anguilla japonica Temminck &
Schle-gel and Anguilla marmorata Quoy & Gaimard) (Shiao, 1998; Liu et al., 2000; K. N. Shen, Y. C. Lee & W. N. Tzeng, unpubl. data; C. S. Tzeng, unpubl. data).
Sicyopterus japonicus is one of the few amphidromous gobies that can migrate
far inland (Liu et al., 2000). Some individuals can be found 80 km upstream from the estuary in the Hsuikuluan River (C. S. Tzeng & J. C. Shiao, unpubl. data).
In eastern Taiwan, S. japonicus are only found in some less polluted streams (Shiao, 1998; Liu et al., 2000) and are used as an indicator species for stream pollution because they will only migrate to streams with good water quality (Liu et al., 2000; Wang, 2002). Sicyopterus japonicus is one of the most abundant and dominant amphidromous gobies in these streams. In the Hsuikuluan River, for example, it is the second most dominant goby species and constitutes 21% of the fishes. Therefore, the conservation of habitat is important for the sustainable use of these amphidromous gobies (Liu et al., 2000; Wang, 2002). Amphidro-mous gobies are also found in other subtropical or tropical areas (Manacop, 1953; Radtke et al., 1988; Bell et al., 1995; Radtke & Kinzie III, 1996; Hoareau et al., 2007a, b). Their life histories are similar, but the length of marine larval duration and size at recruitment varies (Bell et al., 1995; Radtke et al., 2001; Hoareau et al., 2007a, b). For example, two Hawaiian amphidromous gobies,
Stenogobius hawaiiensis Watson and Awaous guamensis (Valenciennes), have
a remarkably long pelagic larval duration (119–151 days and 150–169 days, respectively) (Radtke et al., 1988). Sicyopterus lagocephalus (Pallas) is known
to have the longest pelagic larval duration (mean S.D. 199 33 days) and is
distributed over a wide area in the Indo-Pacific (Keith et al., 2005; Hoareau et al., 2007a). In two Dominican gobies, Sicydium punctatum Perugia and Sicydium
anti-llarum Ogilvie-Grant, the pelagic larval stages are long and seasonally inverse
cyclic changes in age and total length (LT) at recruitment occur (Bell et al.,
1995), which means the individuals recruited at a large size are younger and vice versa. The seasonal difference in growth rate may be influenced by environmental factors, such as water temperature and food availability (Yu & Ueng, 2001).
Otoliths of many fishes provide a daily record of previous events, including the duration of the pelagic larval stage (Brothers et al., 1976, 1983; Victor,
1986a; Radtke et al., 1988; Thresher & Brothers, 1989; Wellington & Victor, 1989), settlement timing (Victor, 1982, 1984; Shen & Tzeng, 2002) and relative growth rates during different life-history stages (Victor, 1986b; Thorrold & Mil-icich, 1990; Jones, 1992; Searcy & Sponaugle, 2000). The marine larval stage of
S. japonicushas been confirmed by otolith Sr:Ca ratio analysis (Shen et al., 1998;
Shen & Tzeng, 2002), but little is known about their spawning strategy, early growth history and seasonal recruitment dynamics and how these life-history variables contribute to the abundance of S. japonicus in East Asia. The aim of the present study was to understand not only the pelagic larval duration but also the spawning seasons and recruitment regime of S. japonicus by counting the daily growth increments in the otolith of the postlarvae collected year round from the surf zones at the river mouth during their upstream migration. The sea-sonal variation in the age and size at recruitment and the growth rate of the marine life stage in relation to physical and biological factors were also analysed.
MATERIALS AND METHODS
S A M P L I N G D E S I G N
The Shuang-Chi River is a less polluted river than most rivers in Taiwan in north-eastern Taiwan with a length of c. 268 km and catchment of 1325 km2. The width of the river mouth is c. 50 m. The water from Shuang-Chi River discharges into the Yen-Liao Bay where the outer part of the bay faces the Kuroshio Current some 100 km away from the shore (Tzeng et al., 1997). A total of 879 newly recruited S. japonicus postlarvae were collected with a stationary net set against flood tide in the surf zones at the mouth of the Shuang-Chi River (Fig. 1). Sampling was conducted twice a month (one close to the last quarter and another close to first quarter of the lunar cycle) at night during spring tide from May 1996 to April 1997. Each sampling lasted for 3 h before high tide, collecting the fish entering the river with the tidal current. A sub-sample collected from the by-catch of the glass eel fishery on 22 and 24 February and 11 March 1996 at the same location was also used for otolith analysis. Almost all the samples were used for age determination except those collected on 10 May 1996, where 50 of the 612 individuals collected were ran-domly selected for LTmeasurement and age determination from the major recruitment. The
sampling dates, numbers of individuals from 3 h collections and the number of individuals used for LT measurement and age determination are shown in Table I. Physical factors
including salinity and water temperature were measured with a salinometer (WTW Cond 330i, Weilheim, Germany) at a depth of 05 m during sampling.
O T O L I T H A G E I N G
The sagittal otoliths of the postlarvae were extracted under a stereomicroscope. After cleaning with distilled water and drying in air, the otoliths were mounted with DPX (Fluka, Steinheim, Germany) on a glass slides with a cover slip. The daily growth incre-ment in otoliths could easily be examined with light microscope under200 magnifica-tions without polishing (Fig. 2). The growth increments in the otoliths of S. japonicus are assumed to be deposited on a daily basis similar to other gobies (Radtke et al., 1988; Bell et al., 1995). The number of daily growth increments in each of the otoliths was independently counted twice. If the two readings were different by 1%, the otolith was rejected to reduce the ageing error. To understand if recruitment is size dependent or age dependent, the regression of LTand age at recruitment was calculated. The
coef-ficient of variation (C.V., y) was from y¼ 100sx 1;where x is the sample mean and s is theS.D.
B A C K C A L C U L A T I O N O F H A T C H I N G D A T E S A N D P E R I O D I C R E G R E S S I O N A N A L Y S E S
It is not known when the first increment was deposited in the otolith of S. japonicus, but it was assumed that it was deposited at hatching as in other amphidromous goby species (Bell et al., 1995). No obvious check marks were found in the otoliths of the newly recruited postlarvae because when they arrive in the estuary metamorphosis is not complete (Shen & Tzeng, 2002). The hatching dates of the larvae were backcalcu-lated from the sampling dates and the daily age at recruitment. Meanwhile, a periodic regression curve (Batschelet, 1981) was fitted for the LT and daily age at recruitment
and from these, the growth rate was estimated to understand the seasonal changes in age, LT and the growth rate of the larvae at recruitment. In addition, periodic
regres-sions between LT at recruitment and hatching dates were also calculated (Bell et al.,
1995). The calculation for periodic regression was based on the formula y ¼ M þ A cos (ot ot0), where M is the mean level, A is amplitude (A 0), o is angular
fre-quency and t and t0 are the time and time of peak phase, respectively. The significance
of periodic regression was tested after periodic regression transfer to the linear regres-sion (Batschelet, 1981).
T H E R E L A T I O N S H I P B E T W E E N S E A S O N A L A B U N D A N C E A N D E N V I R O N M E N T A L F A C T O R S
The seasonal abundances of S. japonicus at recruitment during the different sampling dates were log10transformed due to the large variances in the abundance of recruiting
individuals and were plotted against water salinities and temperatures separately. The optimal environmental condition for postlarvae at recruitment was defined as the 95% of the total catch.
FIG. 1. The sampling site ( ) of Sicyopterus japonicus in Shuang-Chi River estuary, north-eastern Taiwan.
RESULTS
A G E A N D S I Z E A T R E C R U I T M E N T
The mean S.D. LT of S. japonicus at recruitment to the Shuang-Chi River
estuary was 3395 131 mm (range 308 to 381 mm, n ¼ 317) with a mean
S.D. age of 16372 1279 days (range 130 to 198 days, n ¼ 312). The LT at
recruitment was significantly and positively correlated with the daily age
(r2 ¼ 01608, P < 0001, n ¼ 312), but the coefficient of determination was
low (Fig. 3). The C.V. of the LT (3%) was less than age (8%).
TABLEI. The sampling date, lunar phases, number of fish collected, numbers of fish used
for total length (LT), daily age measurement and the salinity and water temperature
during sampling Sampling date Lunar phase Number of fish collected The number of fish used for Salinity Temperature (° C) LT (mm) Age (days)
22 February 1996 First quarter 32 32 30 — — 24 February 1996 First quarter 7 7 7 — — 11 March 1996 Last quarter 34 34 34 — — 10 May 1996 Last quarter 612 50 48 140 200 24 May 1996 First quarter 0 0 0 02 210 8 June 1996 Last quarter 7 7 7 45 290 20 June 1996 First quarter 0 0 0 185 265 4 July 1996 Last quarter 0 0 0 325 288 18 July 1996 First quarter 0 0 0 320 288 5 August 1996 Last quarter 0 0 0 120 275 19 August 1996 First quarter 0 0 0 295 265 5 September 1996 Last quarter 3 3 3 90 260 19 September 1996 First quarter 0 0 0 280 250 4 October 1996 Last quarter 3 3 3 80 238 17 October 1996 First quarter 37 37 37 98 230 4 November 1996 Last quarter 0 0 0 09 237 18 November 1996 First quarter 11 11 11 08 190 3 December 1996 Last quarter 0 0 0 240 180 18 December 1996 First quarter 9 9 9 150 163 4 January 1997 Last quarter 29 29 28 20 179 15 January 1997 First quarter 14 14 14 05 163 1 February 1997 Last quarter 0 0 0 105 161 17 February 1997 First quarter 4 4 4 11 162 4 March 1997 Last quarter 15 15 15 10 181 19 March 1997 First quarter 40 40 40 02 185 1 April 1997 Last quarter 22 22 22 319 212
Total 879 317 312
S E A S O N A L C H A N G E S I N T O T A L L E N G T H A N D D A I L Y A G E A T R E C R U I T M E N T A N D S O M A T I C - G R O W T H R A T E
The mean LTat recruitment changed cyclically with sampling date [Fig. 4(a)].
A periodic regression equation of the LT at recruitment on the sampling date
was calculated (r2 ¼ 0280, n ¼ 16, P < 005). The slope of the regression
was significant, indicating that LT at recruitment displayed a yearly cyclic
FIG. 2. Daily growth increments in the otolith of the juvenile Sicyopterus japonicus, 33 mm total length. , primordium. Scale bar¼ 100 mm.
change with the largest size at recruitment appearing on 15 November 1996 and the smallest size at recruitment appearing on 16 May 1996. In other words, the gobies recruited to the estuary were larger in autumn and smaller in spring. The mean daily age at recruitment of the fish also changed cyclically with sampling date [Fig. 4(b)]. A periodic regression equation of the daily age at
re-cruitment on the sampling date was calculated (r2 ¼ 0315, n ¼ 16, P < 005).
The regression was significant, indicating that the age of the fish at recruitment also displayed a yearly cyclic change with the older fish at recruitment appear-ing on 29 June 1996 and the younger fish at recruitment appearappear-ing on 18
December 1996. The peak phase of LT and age at recruitment was inversely
related, the larger fish were younger at recruitment and vice versa.
The average somatic-growth rate changed cyclically with sampling date [Fig. 4(c)]. A periodic regression equation of the somatic-growth rate on sampling date
was calculated (r2 ¼ 0372, n ¼ 16, P < 005). The regression was significant,
indicating that the mean growth rate of the fish at recruitment also displayed a yearly cyclic change with a maximum-somatic growth rate on 8 December 1996 and a minimum-somatic growth rate on 8 June 1996. The seasonal change
in somatic-growth rate was similar to that of LT but opposite to that of age.
T H E R E L A T I O N S H I P B E T W E E N T O T A L L E N G T H A T R E C R U I T M E N T A N D H A T C H I N G D A T E
The duration of the estimated hatching dates of S. japonicus was quite long, almost year around except for the winter (Fig. 5). According to the abundance of the larvae during recruitment in spring, the main spawning season was in
autumn. A cyclic regression of LT at recruitment on hatching date was
calcu-lated (r2¼ 0070, n ¼ 312). The regression was significant (P < 0001),
indicat-ing that the size of the larvae at recruitment changed with the date at hatchindicat-ing, the larger larvae which recruited to the estuary in the autumn [Fig. 4(a)] were spawned on 22 April 1996 (spring), and the smaller ones at recruitment in spring were expected to have been spawned on 22 October 1995 (autumn).
Sicyopterus japonicus spawned in spring were larger at recruitment on the
0 5 10 15 20 25 30 35 40 120 140 160 180 200 Age (days) LT (mm)
FIG. 3. The relationship between total length (LT) and daily age at recruitment of Sicyopterus japonicus. The curve was fitted by: y¼ 00041x þ 27256.
average than those spawned in autumn. The LTwas more divergent, however,
in the spring than the autumn cohort (Fig. 5).
F A C T O R S A F F E C T I N G T H E A B U N D A N C E O F R E C R U I T M E N T
Sicyopterus japonicus postlarval recruitment to the estuary was almost year
round except during summer (July and August) (Fig. 6). Approximately 95%
of the larvae were recruited when the water temperature was23° C [Fig. 6(a)],
which coincided with the annual temperature cycle except during summer and
0·17 0·18 0·19 0·2 0·21 0·22 0·23 140 150 160 170 180 190 31 32 33 34 35 36 37 LT (mm) Age (day)
Growth rate (mm day
–1 ) (a) (b) J F M A M J J A S O N D J F M A (c) 1996 1997 Sampling dates
FIG. 4. Seasonally cyclic changes of meanS.D. (a) total length (LT), (b) age and (c) growth rate at recruitment from February 1996 to April 1997. The cyclic changes were fitted with periodic regression equations: (a) y¼ 3392 þ 037 cos (ox 2385) (b) y ¼ 162953 þ 2500 cos (ox þ 0012) and (c) y¼ 0200 þ 0005 cos (ox 27833) where o ¼ 2pT1(T¼ 365 days). The sample size on each sampling date is the same as shown in Table I.
also 95% of the newly recruited larvae were captured when the salinity was14 [Fig. 6(b)]. When the salinity was too high, no fish were caught even during the recruitment period (19 September and 3 December 1996), but there were three sampling dates (24 May and 4 November 1996 and 1 February 1997) when no fish were collected even though the salinity was low. This may have been due to the high water flow from the river after heavy rains (the accumulated precipi-tation of these 3 months were 3845, 9355 and 3310 mm, respectively).
DISCUSSION
T H E S P A W N I N G S T R A T E G Y O F S . JA P O N I C U S
Backcalculated spawning periods from the otolith daily increments suggest that the spawning of S. japonicus is almost year-round except during winter, with a major event in autumn. Reproductive biology research on S. japonicus in the Jinlun Stream, in south-eastern Taiwan (Ju, 2001) also found that the
gonado-somatic index (IG) of S. japonicus was at a low level from November
to April but with a peak during September and October. It increased from 059 in May to 668 in October and suddenly decreased to 043 in November. The fecundity of S. japonicus is c. 11 000–18 000 and egg size is c. 72–103 mm
diameter (n¼ 3, 64–78 mm LT) (Ju, 2001). This shows that the main spawning
season is in autumn and that the deposition of the growth increments in the otolith of S. japonicus is on a daily basis.
Shuang-Chi River estuary is in a subtropical area where there is a major spring algal bloom and a minor bloom caused by the recycling of the nutrients in the autumn, as in temperate areas (Huang et al., 1985). The spawning season of many fishes coincides with the spring phytoplankton bloom in temperate areas to match the timing of the production cycle (match and mismatch hypothesis; Cushing, 1975). The planktonic foods are critical to the survival, growth and year class strength of many fish species (van der Veer & Witte,
29 31 33 35 37 39 A S O N D J F M A M J J A S O N LT (mm) 1995 1996 Hatching date
FIG. 5. Total length (LT) at recruitment and hatching date of Sicyopterus japonicus during the period from August 1995 to October 1996. , the samples from the main recruitment on 10 May 1996;r, samples from year-round recruitment. The curve was fitted by: y¼ 34119 þ 048 cos (ox þ 1245, where w¼ 2pT1(T¼ 365 days).
1999). Large numbers of fish eggs and larvae appear in spring because of the plentiful food supply (Su et al., 1981; Huang et al., 1985). Copepods, the main food of S. japonicus larvae in the marine larval stage (Dotu & Mito, 1955), are also more abundant in warm than cold seasons in Yen-Liao Bay (Tzeng et al., 1997). Why do the S. japonicus choose the low production season (autumn) as their main spawning season?
The climate in subtropical areas is characterized by a dramatically seasonal change with an intermittent precipitation and riverflow (Chapman & Kramer, 1991) similar to Taiwan. Winter and early spring (December to March) is the dry season in Taiwan, and the water level and temperature are low in the river during this period. The following spring rain may initiate the spawning of some individuals, but not the majority. The seasonal heavy rains brought on by ty-phoons during summer and autumn increases the water volume of small streams in eastern Taiwan. Amphidromous gobies may take this opportunity to spread their larvae downstream and to complete their downstream migration towards the sea, and thus, the spawning season of the fishes may be synchronized
0·0 0·5 1·0 1·5 2·0 2·5 3·0 0 5 10 15 20 25 30 35 0·0 0·5 1·0 1·5 2·0 2·5 3·0 10/5/96 24/5/96 8/6/96 20/6/96 4/7/96 18/7/96 5/8/96 19/8/96 5/9/96 19/9/96 4/10/96 17/10/96 4/11/96 18/11/96 3/12/96 18/12/96 4/1/97 15/1/97 1/2/97 17/2/97 4/3/97 19/3/97 1/4/97 0 5 10 15 20 25 30 35 Log 10 (number of individuals) Sampling dates (a) (b) Salinity Water temperature (° C)
FIG. 6. The log10number of individuals ( ) in relation to (a) water temperature ( ) and (b) salinity ( ). The number of individuals located in salinity <14 and water temperature <23° C is 95%.
with the intermittent precipitation (Ryan, 1991). K. Ego (unpubl. data) found that
A. guamensisin Hawaii spawns from August to December with a peak spawning
during high water following the rainstorms of late summer or autumn. Radtke
et al. (1988) also suggested that Hawaiian gobies spawn during the rainy season
because the larvae may easily be swept downstream to the ocean upon hatching. The positive relationship between spawning period and precipitation was also
demonstrated by the IG of S. japonicus in south-eastern Taiwan (Ju, 2001).
Amphidromous gobies are abundant in tropical and subtropical small streams and rivers of the Indo-Pacific, which have a short length, such as in Japan (Watanabe et al., 2006; Yamasaki et al., 2007), in Taiwan (Shen et al., 1998), in Hawaii (Radtke et al., 1988) and in the Caribbean (Bell et al., 1995). The newly hatched larvae in the short streams spent less time without food prior to reaching the plankton-rich, brackish and marine zone (Iguchi, 2007). Therefore, precipitation plays an important role in speeding the larval dispersal of amphidromous gobies (Erdman, 1961; Delacroix & Champeau, 1992; Fitzsimons et al., 1997; Keith et al., 2006). In addition to precipitation, the seasonal decrease in temperature and day length may also be cues that in-duce fish spawning. For example, in most fish species native to North America, changing day length and water temperature are strong proximate factors that influence the reproductive cycle (Sumpter, 1990).
Bell & Brown (1995) found that a salinity of <10 in an estuary is essential for the survival of 0–5 day-old Sicydiine larvae. The seasonal heavy rain makes the estuary act as a salinity buffer and keeps the larvae away from the deadly high salinity sea water. The adult fishes can detect the cues of these environ-mental changes and take this chance to disperse their offspring and decrease the risk of mass mortality.
T H E G R O W T H H I S T O R Y O F S . JA P O N I C U S I N T H E M A R I N E L A R V A L S T A G E
Information about the early growth history of S. japonicus in the marine lar-val stage is limited, but growth records from otoliths may give some hints. The recruitment period of the S. japonicus from ocean to stream is from September
to June. The peak phase of cyclic change in age and LT of the fish at
recruit-ment is inversely related, and the LTof fish at recruitment is larger in autumn
than in spring. The age of fish at recruitment, however, is younger in autumn than in spring, and fish recruited in autumn and winter were hatched in spring as indicated from the backcalculation of the otolith daily growth increments, while the spring recruited fish were hatched in the autumn of the previous year.
The significant cyclic regression between LT at recruitment and hatching date
also shows a similar pattern, the fish hatched in autumn were smaller at recruit-ment and the fish hatched in spring were larger. The significant difference, however, might be due to the large sample size (Kubinger et al., 2007). On the other hand, the size variation among individuals of the same cohort is a common phenomenon in fishes (DeAngelis et al., 1993; van Densen et al., 1996; Huss et al., 2007), which may also contribute to the low coefficient of determination (Fig. 5). The pelagic larval environment (e.g. temperature, food availability and oceanographic features) has been shown to alter the rates of
larval growth (Hovenkamp & Witte, 1991; Benoıˆt & Pepin, 1999; Otterlei et al., 1999; Keller & Klein-MacPhee, 2000), development time (Benoıˆt & Pepin, 1999; Otterlei et al., 1999; Searcy & Sponaugle, 2000) and survival to settlement (Meekan & Fortier, 1996; van der Veer & Witte, 1999; Keller & Klein-MacPhee, 2000) in a variety of marine fishes.
In Yen-Liao Bay, surface water temperatures were high in August (mean
S.D. 293 02) and low in winter (187 02). The abundance of zooplankton,
fish larvae and eggs and the number of fish species were all higher in spring and fewer in autumn, reaching a low in winter (Tzeng et al., 1997). Therefore, spring-hatched larvae live in a highly productive and high water temperature environment, while autumn-hatched larvae live in a relatively poor productive and low water temperature environment (Su et al., 1981; Huang et al., 1985; Tzeng et al., 1997). In other words, the environment and nutritional condition were different for spring and autumn-hatched larvae. They may even use differ-ent nursery areas as shown in spring and autumn-hatched herring Clupea
hare-ngus L. (Rinne, 1988). During the high productivity and high water
temperature environment in spring, the larvae can grow faster and recruit to the estuary in autumn of the same year, although they may have to face more predation risk and competition with the other species (Warlen & Burke, 1990). This may be the reason why the average size and age of the larvae at recruit-ment were larger and younger in autumn, but with a higher size variation. Conversely, fish that recruited to the estuary in the late spring and hatched in autumn face a relatively poor productivity and low temperature environment (Su et al., 1981; Huang et al., 1985; Tzeng et al., 1997) but also less competitors and predators, which may cause them to grow relatively slowly and thus lead them to be smaller and older at recruitment.
The low coefficient of determination between daily age and LTof S. japonicus
were also found in other amphidromous gobies (Hoareau et al., 2007b). The
low C.V. of the LT relative to age indicated that the recruitment of the S.
japo-nicus was size dependent rather than age dependent. Fernandez-Diaz et al.
(2001) showed that metamorphosis and recruitment of the Senegal sole Solea
senegalensis Kaup is size dependent rather than age dependent. Food quality
was probably a principal factor influencing the timing of metamorphosis from larvae to juvenile. In addition, it was also shown that the larval duration was inversely related to growth rate: faster growing larvae spend less time at the planktonic larval stage (Wellington & Victor, 1989).
Where the S. japonicus larvae stayed during the marine larvae stage is not clear. Smaller marine larvae were seldom collected in the estuary, suggesting that the estuary is not the place for marine larvae. Research in the Yen-Liao Bay area by Tzeng et al. (1997) found the larvae of the amphidromous goby
Rhinogobius sp. 6 km from the river mouth in March 1993. This may be
evi-dence for coastal retention of these amphidromous gobies during the marine larval stage.
R E C R U I T M E N T D Y N A M I C S O F S . JA P O N I C U S
Due to the mountainous features of East Taiwan, the rivers are all small and with a steep gradient. Therefore, the streams in eastern Taiwan are usually less
polluted than those of western Taiwan, and a better environment for size-dependent rather than age-size-dependent amphidromous recruitment. The plank-tonic larval duration of S. japonicus is longer than for most other species (Radtke et al., 1988; Bell et al., 1995; Shen et al., 1998; Yamasaki et al., 2007). Longer larval duration makes them more likely to disperse, but it does not guarantee recruitment success (Maeda et al., 2007). This lengthy larval dura-tion has also been observed in the widely distributed amphidromous S.
lagocepha-lus in Indo-Pacific area (Hoareau et al., 2007a). The LT size of S. japonicus at
recruitment is larger than in other amphidromous goby species in Taiwan such as Rhinogobius gigas Aonuma & Chen, which is only 176 mm at recruitment (Shiao, 1998) and hence liable to be less affected by predators or environmental factors in the estuary (Miller et al., 1988; Bailey & Houde, 1989).
Precipitation and temperature not only play an important role in determin-ing the timdetermin-ing of spawndetermin-ing in amphidromous fishes but also the timdetermin-ing of recruitment of their postlarvae. In this study, c. 95% of the specimens were col-lected at a salinity <14, and seldom colcol-lected at a salinity >14, even during the recruitment season. This suggests that low salinity water may provide a suitable environment for the postlarvae to recruit to streams. The timing of the upstream migration of diadromous freshwater eels, for example, is also strongly correlated with rainfall (Jellyman & Ryan, 1983; Chen-Lee et al., 1994). In addition, salinity is the most important factor guiding the choice of water by glass eels of Anguilla anguilla (L.) (Tosi et al., 1990). In addition, 95% of the specimens in this study were collected at a temperature of <23° C. The absence of recruitment in a high temperature environment during summer in north-eastern Taiwan coincides with the spawning gap of S. japonicus in win-ter, and lunar phases are also a major factor affecting the recruitment of am-phidromous gobies (Delacroix & Champeau, 1992; Shiao, 1998; Keith, 2003; Hoareau et al., 2007b).
In summary, the abundance of S. japonicus in eastern Taiwan and their wide distribution in East Asia indicates that their spawning strategies are successful. These life-history strategies and their relationship to environmental cues are the result of a long-term evolutionary adaptation that coincides with environmental cues that facilitate the dispersal and survival of the larvae by a matching of abiotic and biotic factors. The conservation and management of this species will depend on both this basic life-history information together with habitat conservation.
This study was conducted with the financial support of the National Science Foun-dation, Republic of China (Project No. NSC 096-2811-B-002-030). The authors are grateful to N. J. S. Leander for reading the English text and several anonymous reviewers of an earlier version of this manuscript.
References
Bailey, K. M. & Houde, E. D. (1989). Predation on eggs and larvae of marine fishes and the recruitment problem. Advances in Marine Biology 24, 1–83.
Batschelet, E. (1981). Circular Statistics in Biology. London: Academic Press.
Bell, K. N. I. & Brown, J. A. (1995). Active salinity choice and enhanced swimming endurance in 0 to 8-d-old larvae of diadromous gobies, including Sicydium punctatum(Pisces), in Dominica, West Indies. Marine Biology 121, 409–417.
Bell, K. N. I., Pepin, P. & Brown, J. A. (1995). Seasonal, inverse cycling of length- and age-at-recruitment in the diadromous gobies Sicydium punctatum and Sicydium antillarum in Dominica, West Indies. Canadian Journal of Fisheries and Aquatic Sciences 52,1535–1545.
Benoıˆt, H. P. & Pepin, P. (1999). Individual variability in growth rate and the timing of metamorphosis in yellowtail flounder Pleuronectes ferrugineus. Marine Ecology Progress Series 184,231–244.
Brothers, E. B., Mathews, C. P. & Lasker, R. (1976). Daily growth increments in otoliths from larval and adult fish. Fishery Bulletin 74, 1–8.
Brothers, E. B., Williams, D. M. & Sale, P. F. (1983). Length of larval life in twelve families of fishes at ‘One Tree Lagoon’, Great Barrier Reef, Australia. Marine Biology 6,319–324.
Chapman, L. J. & Kramer, D. L. (1991). The consequences of flooding for the dispersal and fate of poeciliid fish in an intermittent tropical stream. Oecologia 87, 299–306. Chen-Lee, Y. L., Chen, H. Y. & Tzeng, W. N. (1994). Reappraisal of the importance of rainfall in affecting catches of Anguilla japonica elvers in Taiwan. Australian Journal of Marine and Freshwater Research 45,185–190.
Cushing, D. H. (1975). Marine Ecology and Fisheries. Cambridge: Cambridge University Press.
DeAngelis, D. L., Rose, K. A., Crowder, L. B., Marschall, E. A. & Lika, D. (1993). Fish cohort dynamics-application of complementary modeling approaches. American Naturalist 142,604–622.
Delacroix, P. & Champeau, A. (1992). The breeding in freshwater of Sicyopterus lagocephalus (Pallas), a gobiid fish of the Reunion rivers. Hydroecologie Appliquee 4,49–63. van Densen, W. L. T., Ligtvoet, W. & Roozen, R. W. M. (1996). Intra-cohort variance in
the individual size of juvenile pikeperch, Stizostedion lucioperca, and perch, Perca fluviatilis, in relation to the size spectrum of their food items. Annales Zoologici Fennici 33,495–506.
Dotu, Y. & S. Mito (1955). Life history of a gobioid fish, Sicydium japonicum Tanaka. Science Bulletin of the Faculty of Agriculture, Kyushu University 15, 213–221 (in Japanese with English summary).
Erdman, D. S. (1961). Notes on the biology of the gobiid fish Sicydium plumieri in Puerto Rico. Bulletin of Marine Science of the Gulf and Caribbean 11, 448–456.
Fernandez-Diaz, C., Yufera, M., Canavate, J. P., Moyano, F. J., Alarcon, F. J. & Diaz, M. (2001). Growth and physiological changes during metamorphosis of Senegal sole reared in the laboratory. Journal of Fish Biology 58, 1086–1097.
Fitzsimons, J. M., Schoenfuss, H. L. & Schoenfuss, T. C. (1997). Significance of unimpeded flows in limiting the transmission of parasites from exotics to Hawaiian stream fushes. Micronesica 30, 117–125.
Hoareau, T. B., Bosc, P., Valade, P. & Berrebi, P. (2007a). Gene flow and genetic structure of Sicyopterus lagocephalus in the south-western India Ocean, assessed by intron-length polymorphism. Journal of Experimental Marine Biology and Ecology 349,223–234.
Hoareau, T. B., Lecomte-Finiger, R., Grondin, H. P., Conand, C. & Berrebi, P. (2007b). Oceanic larval life of La Reunion‘bichiques’, amphidromous gobiid post-larvae. Marine Ecology Progress Series 333,303–308.
Hovenkamp, F. & Witte, J. I. J. (1991). Growth, otolith growth and RNA/DNA rations of larval plaice Pleuronectes platessa in the Southern Bight of the North Sea 1987– 1989. Marine Ecology Progress Series 58, 201–215.
Huang, C. C., Tzeng, W. N. & Lee, S. C. (1985). Preliminary survey on larval fishes of Yen-Liao Bay, Northeaster Taiwan. Bulletin Institute Zoology, Academia Sinica 24, 147–154.
Huss, M., Persson, L. & Bystro¨m, P. (2007). The origin and development of individual size variation in early pelagic stages of fish. Oecologia 153, 57–67.
Iguchi, K. (2007). Limitations of early seaward migration success in amphidromous fishes. In Biology of Hawaiian Streams and Estuaries (Evenhuis, N. L. & Fitzsimons, J. M., eds), pp. 75–85. Bishop Museum Bulletin in Cultural and Environmental Studies 3.
Jellyman, D. J. & Ryan, C. M. (1983). Seasonal migration of elvers (Anguilla spp.) into lake Pounui, New Zealand, 1974-1978. New Zealand Journal of Marine and Freshwater Research 17,1–15.
Jones, C. M. (1992). Development and application of the otolith increment technique. In: Otolith Microstructure: Rxamination and Analysis(Stevenson, D. K. & Campana, S. E., eds), pp. 1–11. Canadian Special Publication of Fisheries and Aquatic Sciences 117. Ju, Y. M. (2001). Morphological taxonomy and molecular evolution of mtDNA of genus
Sycidiine and the reproduction ecology of Sicyopterus japonicus of Taiwan. MS Thesis, Institute of Marine Resources, University of Sun Yet-Sen, Taiwan. Keith, P. (2003). Biology and ecology of amphidromous Gobiidae of the Indo-Pacific and
the Caribbean regions. Journal of Fish Biology 63, 831–847.
Keith, P., Galewski, T., Cattaneo-Berrebi, G., Hoareau, T. & Berrebi, P. (2005). Ubiquity of Sicyopterus lagocephalus (Teleostei: Gobioidei) and phylogeography of the genus Sicyopterus in the Indo-Pacific area inferred from mitochondrial cyto-chrome b gene. Molecular Phylogenetics and Evolution 37, 721–732.
Keith, P., Lord, C. & Vigneux, E. (2006). In vivo observations on postlarval development of freshwater gobies and eleotrids from French Polynesia and New Caledonia. Ichthyological Exploration of Freshwaters 17,187–191.
Keller, A. A. & Klein-MacPhee, G. K. (2000). Impact of elevated temperature on growth, survival, and trophic dynamics of winter flounder larvae: a mesocosm study. Canadian Journal of Fisheries and Aquatic Sciences 57,2382–2392.
Kubinger, K. D., Rasch, D. & Sˇimecˇkova, M. (2007). Testing a correlation coefficient’s significance: using H0: 0 < r l is preferable to H0: r¼ 0. Psychology Science 49,
74–87.
Liu, C. H., Tang, C. L., Wang, A. & Chern, K. H. (2000). Investigation of freshwater fish resources of Lika Stream, Taitung, Taiwan. Journal of National Taitung Teachers College 11,219–246 (in Chinese with English abstract).
Maeda, K., Yamasaki, N. & Tachihara, K. (2007). Size and age at recruitment and spawning season of sleeper, genus Eleotris (Teleostei: Eleotridae) on Okinawa Island, southern Japan. The Raffles Bulletin of Zoology 14, 199–207.
Manacop, P. R. (1953). The life history and habits of the goby, Sicyopterus extraneus Herre (Anga) Gobiidae, with an account of the goby-fry fishery of Cagayan River, Oriental Misamis. Philippine Journal of Fishes 2, 1–7.
Meekan, M. G. & Fortier, L. (1996). Selection for fast growth during the larval life of Atlantic cod Gadus morhua on the Scotian Shelf. Marine Ecology Progress Series 137, 25–37.
Miller, T. J., Crowder, L. B., Rice, J. A. & Marschall, E. A. (1988). Larval size and recruitment mechanisms in fishes: toward a conceptual framework. Canadian Journal of Fisheries and Aquatic Sciences 45,1657–1670.
Myers, G. S. (1949). Usage of anadromous, catadromous and allied terms for migratory fishes. Copeia 1949, 89–97.
Otterlei, E., Nyhammer, G., Folkvord, A. & Stefansson, S. (1999). Temperature- and size-dependent growth of larval and early juvenile Atlantic cod (Gadus morhua): a comparative study of Norwegian coastal cod and northeast Arctic cod. Canadian Journal of Fisheries and Aquatic Sciences 56,2099–2111.
Radtke, R. L. & Kinzie III, R. A. (1996). Evidence of a marine larval stage in endemic Hawaiian stream gobies from isolated high elevation locations. Transactions American Fisheries Society 125,613–621.
Radtke, R. L., Kinzie III, R. A. & Folsom, S. D. (1988). Age at recruitment of Hawaiian freshwater gobies. Environmental Biology of Fishes 23, 205–213.
Radtke, R. L., Kinzie III, R. A. & Shafer, D. J. (2001). Temporal and spatial variations in length of larval life and size at settlement of the Hawaiian amphidromous goby Lentipes concolor. Journal of Fish Biology 59, 928–938.
Rinne, J. (1988). Physical, chemical and hygienic monitoring of the sea area off Helsinki and Espoo. City of Helsinki Water and Wastewater Authority. Reports of the Water Conservation Laboratory 17,55–84 (in Finnish with English abstract).
Ryan, P. A. (1991). The success of the Gobiidae in tropical Pacific insular streams. New Zealand Journal of Marine and Freshwater Research 18,25–30.
Searcy, S. P. & Sponaugle, S. (2000). Variable larval growth in a coral reef fish. Marine Ecology Progress Series 206,213–226.
Shen, K. N. & Tzeng, W. N. (2002). Formation of a metamorphosis check in otolith of the amphidromous goby Sicyopterus japonicus. Marine Ecology Progress Series 228, 205–211. Shen, K. N., Lee, Y. C. & Tzeng, W. N. (1998). Use of otolith microchemistry to investigate the life history pattern of gobies in a Taiwanese stream. Zoological Studies 37, 322–329. Shiao, J. C. (1998). Early life history and fry resources of amphidromous gobies in Hsuikuluan River. Master Thesis, National Tsing Hua University, Hsinchu, Taiwan (in Chinese with English abstract).
Su, J. C., Hung, T. C., Chiang, Y. M., Tar, T. H., Lin, Y. S., Tzeng, W. N., Yang, R. T. & Chen, J. C. (1981). Environmental and ecological studies on Yen-Liao coastal area. Special Publication Scope, Academia Sinica 12,1–198.
Sumpter, J. P. (1990). General concepts of seasonal reproduction. In Reproductive Seasonality in Teleosts: Environmental Influences (Munro, A. D., Scott, A. P. & Lam, T. J., eds), pp. 13–31. Boca Raton, FL: CRC Press.
Thorrold, S. R. & Milicich, M. J. (1990). Comparison of larval duration and pre- and post-settlement growth in two species of damselfish, Chromis atripectoralis and C. Pomacentrus coelestis(Pisces: Pomacentridae), from the Great Barrier Reef. Marine Biology 105,375–384.
Thresher, R. E. & Brothers, E. B. (1989). Evidence of intra- and interoceanic regional differences in the early life history of reef-associated fishes. Marine Ecology Progress Series 57,187–205.
Tosi, L., Spampanato, A., Sola, C. & Tongiorgi, P. (1990). Relation of water odour, salinity and temperature to ascent of glass-eels, Anguilla anguilla (L.): a laboratory study. Journal of Fish Biology 36, 327–340.
Tzeng, W. N., Wang, Y. T. & Chern, Y. T. (1997). Species composition and distribution of fish larvae in Yenliao Bay, Northeastern Taiwan. Marine Biology 36, 146–158. van der Veer, H. W. & Witte, J. I. J. (1999). Year-class strength of plaice Pleuronectes platessain the Southern Bight of the North Sea: a validation and analysis of the inverse relationship with winter seawater temperature. Marine Ecology Progress Series 184,245–257.
Victor, B. C. (1982). Daily otolith increments and recruitment in two coral-reef wrasses, Thalassoma bifasciatumand Halichoeres bivittatus. Marine Biology 71, 203–208. Victor, B. C. (1984). Coral reef fish larvae: patch size estimation and mixing in the
plankton. Limnology and Oceanography 29, 1116–1119.
Victor, B. C. (1986a). Duration of the planktonic larval stage of one hundred species of Pacific and Atlantic wrasses (family Labridae). Marine Biology 90, 317–326. Victor, B. C. (1986b). Larval settlement and juvenile mortality in a recruitment-limited
coral reef fish population. Ecological Monographs 56, 1435–1460.
Wang, H. C. (2002). Indicator fish species for water quality of freshwater stream in Taiwan. Annual Report of the National Institute of Environmental Analysis (NIEA) Taiwan R.O.C. 9, 207–236 (in Chinese).
Warlen, S. M. & Burke, J. S. (1990). Immigration of larvae of fall/winter spawning marine fishes into a North Carolina estuary. Estuaries 13, 453–461.
Watanabe, S., Iida, M., Kimura, Y., Feunteun, E. & Tsukamoto, K. (2006). Genetic diversity of Sicyopterus japonicus as revealed by mitochondrial DNA sequencing. Coastal Marine Science 30,473–479.
Wellington, G. M. & Victor, B. C. (1989). Planktonic larval duration of one hundred species of Pacific and Atlantic damselfishes (Pomacentridae). Marine Biology 101, 557–567.
Yamasaki, N., Maeda, K. & Tachihara, K. (2007). Pelagic larval duration and morphology at recruitment of Stiphodon percnopterygionus (Gobiidae: Sicydiinae). The Raffles Bulletin of Zoology 14,209–214.
Yu, S. L. & Ueng, P. S. (2001). Relationship between growth rate and water temperature of brown trout. Journal of the Fisheries Society of Taiwan 28, 21–26.
