FISHERIES SCIENCE 2006; 72: 985–994
*Corresponding author: Tel: 886-2-2362-2987. Fax: 886-2-366-1198. Email: hsucc@ntu.edu.tw
Received 8 February 2006. Accepted 13 April 2006.
Reproductive biology of female Pacific bluefin
tuna Thunnus orientalis from south-western
North Pacific Ocean
Kuo-Shu CHEN,1 Paul CRONE2 AND Chien-Chung HSU1*
1Institute of Oceanography, National Taiwan University, PO Box 23-13, Taipei 106, Taiwan, and 2South-west Fisheries Science Center, National Marine Fisheries Service, 8604 La Jolla Shore Drive, La Jolla, CA 93027, USA
ABSTRACT: Pacific bluefin is a highly valuable pelagic species that inhabits a broad range in the North Pacific Ocean. The reproductive biology, especially for the spawning aggregation in the south-western North Pacific Ocean, is not well understood. Thus, a total of 119 paired ovary specimens were collected from the Taiwanese longline fleet during the 1999 fishing season (late April through June) to gain a better understanding of important reproductivity-related stock parameters associated with this species. The following conclusions were made: (i) condition factor decreased from late May to early June; (ii) the sex ratio might be 1:1 for spawners; (iii) the gonadosomatic index stayed at a relatively high level and markedly increased from late May to early June; (iv) histological examination of oocytes indicated that all specimens were sexually mature; (v) spawning activity appeared to start in May and peak in late May to early June; (vi) batch fecundity increased with fork length; and (vii) preliminary esti-mates of spawning frequency between batches ranged 2–4.5 days based on analysis of postovulatory follicles.
KEY WORDS: batch fecundity, oocyte stage, Pacific bluefin tuna, sex ratio, spawning
fre-quency, spawning season, Thunnus orientalis.
INTRODUCTION
Pacific bluefin tuna Thunnus orientalis Temmincks and Schlegel 1844 is a large highly migratory pelagic species mostly found in the North and western South Pacific Oceans.1 This species
repre-sents an important economic resource to a host of countries, including Japan, Taiwan, USA, and Mex-ico. Since the early 1990s, from late April through June each year, small-scale Taiwanese longline ves-sels have targeted bluefin tuna that aggregate and undergo spawning migrations in south-western North Pacific Ocean waters from the Philippines to Taiwan. Historically, annual catches by Taiwanese vessels have ranged approximately 1300–2700 t; total annual catches by all nations have ranged 13 000–25 000 t.2
Biological characteristics of this stock, including population status determinations, have been gen-erally defined by Japanese and American scientists
since the 1960s, and by the International Scientific Committee for Tuna and Tuna-like Species in the North Pacific Ocean (ISC) since the late 1990s.2
However, details regarding critical biological parameters, such as maturity, have not been exam-ined in a systematic manner, which confounds conducting timely evaluations on the status of the stock for management purposes. For example, a thorough database that includes results from spe-cific biological studies is currently not readily avail-able to researchers interested in Pacific bluefin population dynamics.
Members of the stock have been reported to move between the western and eastern North Pacific Ocean; however, spawning activity is more constrained in distribution, and is believed to occur only in the western North Pacific Ocean,1,3–6
especially from waters around the Ryukyu Islands to the Sea of Japan from June to August. Reproduc-tive studies have addressed the Pacific bluefin tuna sex ratio, spawning ground distribution, maturity-at-age, spawning fraction, and fecundity.3,6–9
Recently, Tanaka6 reported spawning season,
spawning fraction, and batch fecundity of Pacific bluefin tuna around the Ryukyu Islands and in the
Sea of Japan. Finally, bluefin tuna have been shown to be multiple spawners.10,11
There is little information available regarding spawning behavior for bluefin tuna in south-western North Pacific Ocean waters (i.e. the area of this study), where the fishery typically harvests very large (mature) adults.12 The primary objectives
of this study are to examine reproductivity-related biological aspects, including spawning season, sex ratio, batch fecundity, and length frequency distri-bution, from critical spawning habitats of bluefin tuna in the North Pacific Ocean.
MATERIALS AND METHODS Sample collection
In total, 241 Pacific bluefin tuna were randomly sampled from 17 different trips of small-scale Tai-wanese longliners (Table 1) that operated in south-eastern waters off Taiwan (Fig. 1) during April and June 1999. Among those, there were 123 females, and 119 ovary specimens were collected onboard the vessels (four ovaries were discarded by fishers). For Pacific bluefin tuna harvests, fish were typically eviscerated immediately onboard the vessel and subsequently chilled. Ovary samples were removed, numbered, and chilled during onboard
procedures, with each sampled fish tagged for later identification associated with measurements (length and weight) collected in an ongoing port-side sampling program in Tungkang. Finally, field-related data regarding the sampled catch included fishing latitude and longitude, as well as other general fishing information gained through inter-views with the captains at the ports following the fishing trip.
Fig. 1 South-western North Pacific where Taiwanese small-scale longline fleet operated to target bluefin tuna and collect female gonad specimens, shaded area.
Table 1 Sampling information and binomial test data for statistical differences in Pacific bluefin tuna sex ratio from fishing-trip data collected from the Taiwanese longline in the south-western North Pacific Ocean in 1999. Null hypo-thesis was based on a sex ratio hypohypo-thesis of 1:1
Trips Dates Females (No.) Males (No.) Total (No.) Probability of female occurrence (P) Test results 1 9 May 7 1 8 0.0313 * 2 17 May 8 4 12 0.1208 ns 3 19 May 2 6 8 0.1094 ns 4 22 May 6 2 8 0.1094 ns 5 23 May 7 7 14 0.2095 ns 6 26 May 8 9 17 0.1855 ns 7 27 May 7 10 17 0.1484 ns 8 27 May 16 14 30 0.1354 ns 9 27 May 6 6 12 0.2256 ns 10 31 May 6 10 16 0.1222 ns 11 31 May 6 6 12 0.2256 ns 12 01 June 7 9 16 0.1746 ns 13 03 June 4 7 11 0.1611 ns 14 06 June 14 12 26 0.1439 ns 15 08 June 6 3 9 0.1641 ns 16 12 June 10 9 19 0.1762 ns 17 13 June 3 3 6 0.3125 ns Sum 123 118 241 0.0487 * Sum† 116 117 233 0.0521 ns
ns, not significant (P > 0.05); *significant (P < 0.05).
†
Ovary samples were immediately transported from the landing site to the laboratory, wiped clean of moisture, and weighed to the nearest 20 g. Small pieces (∼50 g) were subsampled from the anterior, middle, and posterior regions of both lobes of each ovary (Fig. 2), and subsequently fixed in 10% seawater-diluted formalin for further histological examination and analysis.
Morphological descriptions
Since 1993, nearly all fish commercially harvested by the Taiwan longline fishery that operates in the south-western North Pacific Ocean have been measured for fork length (FL). Hsu et al.12
pre-sented length distributions associated with this fishery until 1998, and updated length distribu-tions sampled after 1999 are reported here.
To investigate indices of condition of sampled fish, the following classical Fulton-type condition factor (C ) was used:
(1) where W is eviscerated weight (kg) and L is fork length (FL) (cm).
Sex ratio
Sex ratio was calculated as the proportion of females in all samples collected in this study (Table 1). Under the assumption that vulnerability (probability of capture) to the fishery is equal between sexes, a simple binomial distribution test was conducted to statistically evaluate sex ratio estimates from the sampled fishing trips. If the test generated non-significant findings among trips (i.e. failed to reject the null hypothesis), the sam-ples were pooled and the test repeated on the total, combined sample.
Analysis of ovaries
The gonadosomatic index I was calculated follow-ing Yabe et al.3
(2) where Wov is ovary weight (g) and L is FL (cm).
The developmental stages of oocytes of Pacific bluefin tuna were classified by histological exami-nation. First, to examine the relationship between diameter and developmental stage of oocytes, a 0.03-g sample was removed from each lobe of a sampled ovary, and immersed in a 33% glycerin solution for 10–20 min13 to maintain the equal
pressure inside and outside the eggs. Second, oocytes were detached from ovarian tissue by needles, and diameter (random axis) measure-ments were taken to the nearest 18 μm at 40× mag-nification using a dissecting microscope with an ocular micrometer. A representative diameter for non-spherical oocytes was calculated as the mean of short and long axes.14–16 Oocytes smaller than
0.2 mm were not measured.
Ovaries with hydrated oocytes were randomly selected to evaluate the extent to which these oocytes were homogeneously distributed within an ovary. A nested, one-way analysis of variance (ANOVA) was used to evaluate differences in num-bers of hydrated oocytes among different layers and locations within an ovary.
Estimates of batch fecundity were derived from the ovary sample using the gravimetric method:17 a
cork borer (5 mm) was used to obtain a core
sam-C W L = LL3 ¥ 5 10 ,5 I W L ov W W = ov ¥ 3 L L 4 10 ,4
Fig. 2 Schematic of Pacific bluefin tuna ovary, showing the location (anterior, middle and posterior) and layer (central, mid and outer) where samples were taken.
ple that included tissue from the outer surface of the ovary to the centrally located lumen. The core was further partitioned into the three areas associ-ated with the lumen (outer, middle, and central layers). Separate cores were taken from the ante-rior, middle, and posterior regions of both lobes of each ovary and thus, 18 locations were sampled from the ovary of each fish (Fig. 2). Four subsam-ples (0.03 g) of ovarian tissue were taken from each of the 18 locations within the ovaries and weighed to the nearest 1 mg. Finally, the diameters and numbers of oocytes present in advanced size mode were recorded.
Spawning frequency can be estimated as the inverse of the ratio of spawning females, which is defined as the number of females with identifiable postovulatory follicles divided by the total number of mature females.
The spawning season was determined by exam-ining the spatio-temporal distribution of spawners sampled with the most advanced group of oocyte (MAGOs), and examination of postovulatory folli-cles associated with the ovary samples.
RESULTS
Morphological aspects
Sex-combined FL distributions from 1999 to 2004 illustrated that most fish from this fishing area were larger than 180 cm (Fig. 3), indicating that the fishery harvests almost mature fish. Examining the samples collected in 1999, the condition factor of females decreased slightly from the beginning of
the fishing season in mid-April to the end of the fishing season in early July (Fig. 4). The decline in condition factor might result from increased repro-ductive effort of spawning activity.
Sex ratio
In total, 241 fish were randomly sampled from 17 fishing trips operating in the study area in the 1999
Fig. 3 Length frequency distri-butions for Pacific bluefin tuna caught from the south-western North Pacific by Taiwanese small-scaled longliners, 1999–2004.
Fig. 4 Changes in mean condition factor of female bluefin tuna from the south-western North Pacific by trimester periods from mid-April to early July in 1999; observed data (䊉) and means during trimester period (䉱); numbers above bars are sample sizes.
Condition f actor 3.0 2.5 2.0 1.5 1.0 Mid-April Late-AprilEarl y Ma y Mid-Ma y Late Ma y Trimester period Earl y June
Mid-June Late June Earl y Jul
fishing season, for purposes of sex determination (Table 1). Based on a binomial test, the null hypothesis that the expected probability of female occurrence was 0.5 (50%) within each fishing trip failed to be rejected, i.e. no significant difference in sex ratio was found within fishing trips (P > 0.05). However, pooled fishing trip data across all sam-ples resulted in a significant (P = 0.0487) difference from the hypothesized 1:1 sex ratio (Table 1). More-over, the first trip showed different results from the other 16 trips. Hence, if excluding data obtained in the first trip, the test resulted in a non-significant difference from the hypothesized 1:1 ratio.
Analysis of ovaries
Weights of 119 sampled ovaries ranged 1.7–19.4 kg. Ovary weight generally increased with length (Fig. 5). The estimated gonadosomatic indices were greater than 2.1, with a high value of 18.7 for one fish (Fig. 6). In general, the gonadosomatic indices stayed consistent over the study period, with a slight increase from early in the fishing sea-son to later in the seasea-son.
The developmental process of oocytes for Pacific bluefin tuna were classified into seven stages based on histological examination and diameter mea-surements: (i) chromatin nucleolus stage oocytes (Fig. 7a) were strongly basophilic, with one larger and several smaller nucleoli in the nucleus (oocyte diameters <56 μm); (ii) perinucleolus stage oocytes (Fig. 7b) with several small nucleoli in the periph-ery of the nucleus (diameter 37–131 μm); (iii) yolk vesicle stage oocytes (Fig. 7c) with yolk vesicles in the cytoplasm (diameter 112–243 μm); (iv) early
yolk stage oocytes (Fig. 7d) with small yolk gran-ules in the periphery (diameter 318–336 μm); (v) advanced yolk stage oocytes (Fig. 7e) with large yolk granules distributed in the cytoplasm (dia-meter 375–598 μm); (vi) migratory nucleus stage oocytes (Fig. 7f) had small lipid droplets fused to each other to form larger lipid droplets, with a nucleus that had migrated to the periphery (dia-meter 616–655 μm); and (vii) mature or hydrated stage oocytes (Fig. 7g) that were irregular in shape because of dehydration and had a single, large oil droplet (diameter 673–991 μm). Finally, postovula-tory follicles from a recently spawned female are illustrated in Figure 7h.
Three paired ripe ovaries were subsampled (Fig. 2) for enumerating most advanced group oocytes (MAGOs). A homogeneity test based on ANOVA revealed no significant (P > 0.05) findings for numbers of MAGOs among locations or layers, i.e. the null hypothesis that MAGO numbers were equal across locations and layers of an ovary was rejected (Table 2). Significant (P < 0.001) differ-ences were found between fish used in this analy-sis. Therefore, it was concluded that hydrated oocytes were uniformly distributed within ovaries; however, levels of MAGOs were not statistically similar between the sampled fish.
Oocytes at different stages were found simulta-neously in the hydrated ovaries, with no large hiatus present between non-yolk and early yolk oocytes (Fig. 8). Thus, the development of oocytes was asynchronous and the annual fecundity was indeterminate.
Most specimens were characterized by ovaries with advanced yolk and advanced-developed
Fig. 5 Ovary weight vs fork length of bluefin tuna from the south-western North Pacific during 1999. Ovaries in early yolk stage (䉱), ovaries in advanced yolk stage (䊉), ovaries in migratory nucleus stage (*), and ovaries in mature stage (䊊). The line represents gonadosomatic index = 2. 0 2 4 6 8 10 12 14 16 18 20 170 175 180 185 190 195 200 205 210 215 220 225 230 235 240 245 250 Fork length (cm) Ovary weight (kg)
Fig. 6 Seasonal changes in gonadosomatic index (GSI) of female bluefin tuna from the south-western North Pacific by trimester periods from May to June in 1999; observed data (䊉) and means during trimester period (䉱); numbers above bars are sample sizes.
oocytes, and only a few females with the most advanced group of oocytes in early yolk stage were found in mid-May and early June. As indicated in Figure 9, the proportion of females with ovaries defined under the migratory nucleus stage and mature stage was approximately 15–25% of the total mature females included in this study.
Batch fecundity and spawning frequency
The simple linear relationship between batch fecundity F and fork length L (cm) is illustrated in Figure 10:
F = 3.2393 × 105× L − 5.2057 × 107
(R2= 0.600, n = 20) (3)
Fractions of spawned females in trimester periods, estimated by histological examination of postovu-latory follicles, were relatively high in mid-May
Fig. 7 Oocyte development of bluefin tuna from the south-western North Pacific. (a) Chromatin nucleolus, (b) perinu-cleolus, (c) yolk vesicle, (d) early yolk, (e) advanced yolk, (f ) migra-tory nucleus, and (g) mature or hydrated stages, and (h) postovu-latory follicles shown in spent ovary. Mayer’s hematoxylin– eosin stain. Bar = 0.1 mm. ld, lipid droplet; nl, nucleolus; n, nucleus; od, oil droplet; POs, pos-tovulatory follicles; yv, yolk vesi-cle; yg, yolk granule.
(a) (b) (c)
(d) (e) (f)
(g) (h)
Table 2 ANOVA result of homogeneity test for the number of the migratory nucleus and hydrated oocytes (MAGO) in different regions of the ovaries for three Pacific bluefin tuna Thunnus orientalis sampled from south-western North Pacific
Source of variation d.f. SS MS F P
Fish 2 10 747.24 5373.62 412.82** <0.001
Location 5 88.38 17.68 1.36ns 0.248
Layer (Fish × Location) 10 183.65 18.37 1.41ns 0.188
Error 90 1 171.50 13.02
Total 107 12 190.77
Lobe (right and left); Location (anterior, middle and posterior); Layer (outer, mid-layer and center). **significant (P < 0.01).
ns, not significant (P > 0.05).
Fig. 8 Frequency distribution of oocyte diameter in an ovary with hydrated oocytes of Pacific bluefin tuna. Oocytes <0.2 mm were measured, n = 1668. Specimen was collected on 6 June 6 1999 with 196 cm fork length, 126 kg eviscerated weight and 8260 g ovary wet weight.
(0.50) and mid-June (0.40), with an overall mean estimate of 0.3 (Table 3). For determining esti-mates of spawning frequency from the spawning fractions, it was assumed that postovulatory folli-cles existed approximately 24 h following ovula-tion, postovulatory follicles could be identified accurately, and that sampling of mature females was random. Consequently, the spawning fre-quency of Pacific bluefin tuna in the south-western North Pacific Ocean estimated by the postovula-tory follicle method ranged from 2.0–4.5 days, with a mean of 3.33 days.
Spawning season
Areas where ovaries with hydrated oocytes and postovulatory follicles were found simultaneously in the same ovary are illustrated in Figure 11. High
Fig. 9 Trimester proportion of five developed oocyte stages of bluefin tuna from the south-western North Pacific during the fishing season from May to June, 1999. EYS, early yolk stage; AYS, advanced yolk stage; MNS, migratory nucleus stage; and MS, mature stage. Num-bers above bars are sample sizes.
Table 3 Fraction of females with postovulatory follicles by trimester for bluefin tuna sampled from the waters of south-western North Pacific during 1999 fishing season by Taiwanese small-scaled longline fishery
Time intervals Number of females examined A Females with postovulatory follicles B Fraction of spawned females C = B/A spawning frequency (days) D = 1/C Early May 6 2 0.33 3.0 Mid- May 12 6 0.50 2.0 Late May 27 6 0.22 4.5 Early June 18 4 0.22 4.5 Mid-June 10 4 0.40 2.5
Fig. 10 Relationship between batch fecundity and fork length for bluefin tuna from the south-western North Pacific, n = 20. The line represents the regression of batch fecundity and fork length.
Fig. 11 Areas where ovaries with hydrated (䊉) and spent follicles (䊊) were found during the fishing season.
proportions of ovaries with hydrated oocytes occurred in mid-May and early June, which is con-sistent with the period of high gonadosomatic indices and the presence of the postovulatory follicles. Accordingly, in the waters of the south-western North Pacific Ocean, the fishing season of Pacific bluefin tuna varies annually and thus, the 1999 season may have started earlier than May and peaked from mid-May to early June. This trend is consistent with that observed in condition factor estimates (Fig. 4).
DISCUSSION
Size-frequency and age-frequency distributions
As defined by Rivas,18 Atlantic bluefin tuna Thunnus thynnus individuals are classified as giant
members of the population if FL exceed 180 cm. Thus, Pacific bluefin tuna landed at domestic ports in Taiwan would nearly always be considered giant specimens (Fig. 3). Further, even an individual with a FL of 172 cm analyzed in this study had an oocyte diameter of 556 μm and was determined to be mature. Therefore, it is likely that nearly all Pacific bluefin landings from this Taiwanese lon-gline fishery are composed of sexually mature (giant) fish (Fig. 3).
Based on a generally accepted aging method (length-at-age model),19 the smallest individual
measured in 1999 was 165 cm in FL (Fig. 3), which would translate to an approximately 6-year-old fish. This size may not represent the size at first maturity (length that fish is sexually mature); how-ever, it is the smallest length class of fish observed in the landings since 1999. Further, nearly all of the historical catch by the Taiwanese longline fleet is greater than 190 cm (8-year-old fish), with a mode at 208–219 cm (9–10-year-old fish). Hirota et al.8
and Kumai11 reported that the age of first spawning
of reared Pacific bluefin tuna was 5 years. Tanaka6
reported that smaller Pacific bluefin tuna caught around the Ryukyu Islands and in the Sea of Japan had hydrated oocytes. Joint studies of the mini-mum maturation size and the proportion of mature females at length classes are necessary among international scientists interested in this tuna population.
Sex ratio
The size-combined sex ratio of Pacific bluefin tuna caught in the south-western North Pacific Ocean may not deviate from the expected 1:1 ratio during the period of spawning aggregation, which is
gen-erally consistent with previous reports.7 Therefore,
the occurrence of females seems to be not out-numbered during spawning season. Regarding to the dominance of males by size classes, especially for fish larger than 240 cm FL,20 our results are
inconclusive. Maguire and Hurlbut20 assumed the
predominance of males in fish larger than 242 cm for Pacific bluefin tuna, but this was not the case in the present study.
Batch fecundity and spawning frequency
This study indicated batch fecundity increased with length. Tanaka6 examined batch fecundity
for smaller fish (159–215 cm FL) around Ryukyu Islands and in the Sea of Japan, and found that batch fecundity was also correlated with body size. The aim of our study was to equate the relationship between batch fecundity and FL in the range of 195–240 cm. Moreover, Tanaka’s6 contribution is
important to complete the relationship within an entire range of mature sizes.
The condition factor declined during spawning because of reduced feeding activities. In the present study, the spawning season may be inferred from the condition factor of Pacific bluefin tuna as it decreased during the fishing season in the south-western North Pacific Ocean (Fig. 4). Western Atlantic bluefin tuna either cease or decrease feeding activity during the spawning sea-son.18 Further, fishers experience that ovaries with
hydrated oocytes are rarely found in Pacific bluefin tuna. A reason for this phenomenon is that hydrated oocytes may be present in ovaries for a very short period, and/or fish release oocytes due to stress associated with hooking operations by the fishing fleet. If so, postovulatory follicles may be found in caught fish. Therefore, conclusions drawn from estimates of spawning frequency should be interpreted with caution, given potential biases associated with sampling design and methods used in histological examination. The spawning frequency may be shorter than estimated in Table 3 of the present study.
Spawning season
Based on this study, the spawning season of blue-fin tuna in the south-western North Pacific Ocean is likely from late April through mid-June. Periods when a high proportion of females had hydrated oocytes coincided with occurrence of postovula-tory follicles in an ovary. The proportion of ovaries with postovulatory follicles did not increase across
the spawning season. It appears plausible that the spawning activity moves in a north direction, as advanced yolk oocytes larger than 467 μm were found in ovaries with postovulatory follicles. McPherson21 found that the secondary yolk
oocytes (advanced yolk stage) of yellowfin tuna
Thunnus albacares develop to hydrated oocytes
within 24 h. Further, the oocytes in the migratory nucleus stage of chub mackerel Scomber japonicus suggested spawning occurred within 24 h.22
There-fore, it was concluded that Pacific bluefin tuna with advanced yolk oocytes may spawn within a short period, and the estimated spawning fre-quency across batches may span a few days.
The duration of high gonadosomatic index in Pacific bluefin tuna coincided with estimates of spawning season concluded from this study. More-over, abundant bluefin tuna larvae have been observed in the western North Pacific Ocean dur-ing early May and mid-June,3 as well as in waters
off Taiwan during March and August.23 The spatial
and temporal variabilities in egg distribution documented23 in this study may be due to biases
associated with either the sampling design or larvae identification.
Environmental (oceanographic) conditions influence spawning behavior of tuna.24 The
spawning of Pacific bluefin tuna in the Sea of Japan is reported from late June to August,6 which
is later than that reported in this study. Abun-dant Pacific bluefin tuna larvae occur in May in Taiwanese waters, and the distribution of larvae was correlated with a sea surface temperature of 27°C6,25 and 34–35 salinity.25 Yellowfin tuna in the
Coral Sea spawn when the sea surface tempera-ture is equal to or greater than 26°C.22 North
Atlantic albacore T. allege spawn in a similar range of sea surface temperatures.24 The sea
sur-face temperatures recorded in this study were between 26 and 29°C during April and June. Thus, the spawning of Pacific bluefin tuna is likely related to critical oceanographic parameters such as sea surface water temperature.6 It is likely that
Pacific bluefin tuna in the south-western North Pacific move in a northward direction during spawning periods, eventually reaching waters around Japan.
ACKNOWLEDGMENTS
Thanks to the National Science Council of the Taiwan Government for grants to Dr. Hsu (contract nos. NSC89-2313-B-002-228, NSC90-2313-B002-259 and NSC91-2313-B002-284). Appreciation is extended to numerous fishers who helped to col-lect and measure samples.
REFERENCES
1. Bailiff WH. Synopsis of biological data on the northern bluefin tuna, Thunnus thynnus (Linnaeus, 1758), in the Pacific Ocean. Inter-Am. Trop. Tuna Comm. Spec. Rep. 1980; 2: 261–293.
2. Pacific Bluefin Tuna Working Group. Report of the 3rd ISC
Pacific Bluefin Tuna Working Group, 26–28 January, 2004.
International Scientific Committee on Stock Assessment of Tunas and Tuna-like Species in the North Pacific Ocean (ISC), Honolulu, HI. 2004; 17.
3. Yabe H, Ueyanagi S, Watanabe H. Studies on the early life history of bluefin tuna Thunnus thynnus and on the larvae of the southern bluefin tuna T. maccoyii. Nankai Reg. Fish.
Res. Lab. Rep. 1966; 23: 95–129 (in Japanese).
4. Okiyama M. Occurrence of the postlarvae of bluefin tuna,
Thunnus thynnus, in the Japan Sea. Jpn. Sea Reg. Fish. Res. Lab. Bull. 1974; 25: 89–97.
5. Deriso RB, Bayliff WH (eds). World meeting on stock assess-ment of bluefin tunas: strengths and weaknesses.
Inter-Amer. Trop. Tuna Comm. Spec. Rep. 1991; 7: 357.
6. Tanaka S. Spawning season of Pacific bluefin tuna esti-mated from histological examination of the ovary. A
NRIFSF/IATTC joint workshop on Pacific northern bluefin tuna, in Shimizu, Japan. ISC, Honolulu, HI. 1999; 11.
7. Nakamura H. Preliminary report on the habit of the bluefin, Thunnus orientalis. Zool. Mag. Tokyo 1938; 50: 279–281.
8. Hirota H, Morita M, Taniguchi N. An instance of the matu-ration of 3 full years old bluefin tuna cultured in the floating net. Bull. Japan. Soc. Sci. Fish. 1976; 42: 939.
9. Foreman TJ, Ishizuka Y. Giant bluefin off Southern Califor-nia, with a new California size record. Calif. Fish Game. 1990; 76: 181–186.
10. Baglin RE Jr. Reproductive biology of western Atlantic blue-fin tuna. Fish. Bull. US 1982; 80: 121–134.
11. Kumai H. Studies on bluefin tuna artificial hatching, rearing and reproduction. Nippon Suisan Gakkaishi 1998; 64: 601– 605 (in Japanese).
12. Hsu CC, Liu HC, Wu CL, Huang ST, Liao HK. New informa-tion on age composiinforma-tion and length-weight relainforma-tionship of bluefin tuna, Thunnus thynnus, in the southwestern North Pacific. Fish. Sci. 2000; 66: 485–493.
13. Hunter JR, Lo NCH, Leong RJH. Batch fecundity in multiple spawning fishes. In: Lasker R (ed.). An egg production method for estimating spawning biomass of pelagic fish: Application to the northern anchovy, Engraulis mordax.
NOAA Tech. Rep. NMFS. 1985; 36: 67–77.
14. Liu HC, Su MS. Maturity and Spawning of golden thread (Nemipterus virgatus) from the northern area of the South China Sea. J. Fish. Soc. Taiwan. 1972; 1: 39–46.
15. Schaefer KM. Reproductive biology of black skipjack,
Euthynnus lineatus, an eastern Pacific tuna. Inter-Am. Trop. Tuna Comm. Bull. 1987; 19: 169–260.
16. Chen KS. Studies on reproductive biology of northern blue-fin tuna (Thunnus thynnus) from the waters of Southwest-ern North Pacific. MSc Thesis. Institute of Oceanography, National Taiwan University, Taipei. 2000 (in Chinese with English abstract).
17. Chang KH, Lee CA, Lin C. Study on maturity of the red scad,
Decapterus kurroides aka-adsi ABE in the waters of Taiwan. J. Fish. Soc. Taiwan 1972; 1: 1–9.
18. Rivas LR. A preliminary report on the spawning of the west-ern North Atlantic bluefin tuna (Thunnus thynnus) in the Straits of Florida. Bull. Mar. Sci. Gulf. Caribb. 1954; 4: 302– 322.
19. Yukinawa M, Yabuta Y. Age and growth of the bluefin tuna,
Thunnus thynnus (Linnaeus), in the north Pacific Ocean. Nankai Reg. Fish. Res. Lab. Rep. 1967; 25: 1–18.
20. Maguire JJ, Hurlbut TR. Bluefin tuna sex proportion at length in the Canadian samples 1974–1983. ICCAT, Coll. Vol.
Sci. Pap 1984; 20: 341–346.
21. McPherson GR. Reproductive biology of yellowfin tuna in the eastern Australian Fishing Zone, with special reference to the north-western Coral Sea. Aust. J. Mar. Freshwater Res. 1991; 42: 465–477.
22. Yamada T, Aoki I, Mitani I. Spawning time, spawning fre-quency and fecundity of Japanese chub mackerel, Scomber
japonicus, in the waters around the Izu Islands. Jpn. Fish. Res. 1998; 38: 83–89.
23. Chiu TS, Chen CS. Distribution of Scombrid Larvae (Pisces: Scombridae) in the waters around Taiwan. J. Fish. Soc.
Taiwan. 1995; 22: 303–312.
24. Santiago DJ. Dinamica de la Poblacion de Atun Blanco (Thunnus alalunga Bonaterre 1788) del Atlantico Norte. PhD Thesis. Universidad del Pais Vasco, Bilbao. 2004; 354 (in Spanish).
25. Tan TH, Chen SC. On relation of occurrence of tuna fish larvae to the aquatic environments in adjacent waters of Taiwan and the South China Sea. Acta Oceanogr. Taiwanica 1975; 5: 179–200.
