Marine Science

ICES Journal of

Marine Science ICES Journal of Marine Science (2012), 69(4), 498 –507. doi:10.1093/icesjms/fss037

Oocyte development and maturity classification of boarfish (Capros aper) in the Northeast Atlantic Edward D. Farrell 1*, Karin Hu¨ssy 2, Julie O. Coad2, Lotte W. Clausen 2, and Maurice W. Clarke 1 1

Marine Institute, Rinville, Oranmore, Co. Galway, Ireland National Institute of Aquatic Resources, Technical University of Denmark, Jaegersborg Alle 1, 2920, Charlottenlund, Denmark

2

*Corresponding author: tel: +353-91-387200; fax: +353-91-387201; e-mail: [email protected]

Received 27 May 2011; accepted 2 February 2012.

This study presents the first detailed investigation of the oocyte development and maturity classification of boarfish, Capros aper, which has recently become the target of an industrial fishery in the Northeast Atlantic. A total of 2014 boarfish were collected from January to December 2010. Mature male and female boarfish were sexually dimorphic and could be readily identified based on external characteristics. A comprehensive maturity scale was developed, which indicated that the length at 50% maturity for males and females was 9.7 cm total length. Female boarfish were observed to spawn in Irish waters in June and July. Once spawning ceased the remaining mature oocytes were resorbed. Preliminary analysis of reproductive strategy indicates that the boarfish is likely an asynchronous batch spawner with indeterminate fecundity. Keywords: asynchronous batch spawner, Capros aper, histology, length-at-maturity, oocyte development.

Introduction The boarfish (Capros aper, Linnaeus) is a deep bodied, laterally compressed, pelagic shoaling species distributed from Norway to Senegal, including the Mediterranean, Azores, Canaries, Madeira, and Great Meteor Seamount (Holgersen, 1954; Que´ro, 1986). The abundance of boarfish in the Northeast Atlantic increased exponentially from 1973 to 2002, which was attributed to enhanced recruitment due to an increase in water temperature during the spawning season (Blanchard and Vandermeirsch, 2005). Until recently, the species has been a periodical unwelcome bycatch in mixed demersal, pelagic, and crustacean-trawl fisheries (Fonseca et al., 2005; Borges et al., 2008). However, since 2007, an important industrial fishery has developed, which is executed by Irish, Danish, and Scottish Refrigerated Sea Water (RSW) tank vessels, primarily in ICES Areas VIIg, VIIh, and VIIj. Understanding the reproductive biology of a species is a critical aspect of providing sound scientific advice for fisheries management, as reproductive biology will largely determine the productivity of a stock and thus its resilience to exploitation (Morgan, 2008). Little is known about the reproductive biology of boarfish, though they have been reported to reproduce in the waters southwest of Ireland from June to August (Que´ro, 1986). Some historical studies also noted the occurrence of egg and larval stages during summer (Cunningham, 1888, 1889; Holt and Scott, 1898; Ehrenbaum, 1905; Hefford, 1910). More recently, the

age-at-maturity of boarfish was estimated to be 2 years in the Mediterranean and 5.25 years in the Northeast Atlantic (Kaya ¨ zaydin, 1996; White et al., 2011); however, neither study and O provided a detailed description of maturity estimation and no species-specific maturity scales were developed. To date, there have been no comprehensive studies of the oocyte development or maturity classification of boarfish. These data are essential to form a basis for future studies on the reproductive biology of the species. Evaluation of reproductive strategy is essential for quantifying the reproductive potential of fish species and for identifying suitable methodologies to determine fecundity (Murua et al., 1998; Armstrong and Witthames, 2012). Two modes of fecundity have been described based on the method by which oocytes are recruited to the stock of mature oocytes (Hunter et al., 1992). In species with determinate fecundity, no new oocytes are recruited to the vitellogenic stock during the spawning season; therefore, the potential annual fecundity is equal to the total fecundity before spawning. Conversely, in species with indeterminate fecundity, pre-vitellogenic oocytes are recruited to the stock of vitellogenic oocytes during the spawning season and thus the potential annual fecundity is not fixed before spawning (Murua and Saborido-Rey, 2003; Gordo et al., 2008). In-depth knowledge of reproductive biology is essential for stock assessment and subsequently for the development of a

# 2012 International Council for the Exploration of the Sea. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]

Downloaded from http://icesjms.oxfordjournals.org/ by guest on November 7, 2014

Farrell, E. D., Hu¨ssy, K., Coad, J. O., Clausen, L. W., and Clarke, M. W. 2012. Oocyte development and maturity classification of boarfish (Capros aper) in the Northeast Atlantic. – ICES Journal of Marine Science, 69: 498– 507.

499

Oocyte development and maturity classification of boarfish Table 1. Capros aper specimens collected per sex, length range, and month. Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep

Oct

Nov

Dec

Total

– – – 2 2 3 7 6 7 2 – – 29

– – – 5 10 12 10 9 2 – – 48

– – – 2 5 17 19 21 19 7 1 – 91

– 1 – 2 4 24 30 22 18 20 5 1 127

– – – 2 11 21 23 22 27 31 19 1 157

– 1 7 12 13 25 31 18 30 16 2 – 155

– – – 1 3 12 17 16 15 6 1 – 71

– – – – – – – – – – – – 0

– – 9 10 5 8 11 8 12 4 0 – 67

– – 1 4 6 18 17 14 11 13 3 1 88

6 12 18 5 2 8 14 12 15 6 2 – 100

3 8 20 15 1 6 17 15 24 17 6 1 133

9 22 55 55 57 152 198 164 187 124 39 4 1 066

– – – 2 2 7 6 2 5 3 – 27

– – – – 2 11 14 10 9 2 – 48

– – – – 1 12 18 23 14 5 – 73

– – – – 23 32 23 22 13 1 114

– – – – 6 13 23 37 25 13 2 119

– – – 10 12 25 18 15 15 7 1 103

– – – – – 14 10 6 6 2 – 38

– – – – – – – – – – – 0

– – – 9 22 14 12 9 2 3 71

– – – – 7 20 13 3 10 8 2 63

– 8 10 13 7 10 20 9 13 3 – 93

1 10 5 19 5 – 16 7 13 8 – 84

1 18 15 53 64 149 182 144 134 67 6 833

– – – – – – – – –

– – – – – – – – –

– – – – – – – – –

– – – – – – – – –

– – – – – – 1 – 1

– – – – – – 5 – 5

– – – – – – – – –

– – – – – – – – –

– – – – 1 – 3 – 4

– – – – 1 – – – 1

2 – – 7 20 10 6 1 46

11 3 2 14 16 7 5 – 58

13 3 2 21 38 17 20 1 115

sound and well-founded management plan. To this end, the specific objectives of the current study were to (i) investigate oocyte development, (ii) establish a comprehensive histologically confirmed maturity scale, (iii) accurately determine the length-at-maturity, and (iv) investigate the reproductive strategy.

Material and methods

each comprised two lobes. The ovaries were joined over two-thirds of their length and could not be separated. Testes were more loosely joined; however, separation was difficult without damaging the structure; therefore, they were also left intact. Ovaries and testes were weighed (WOT) to the nearest 0.01 g. Conversion factors between frozen and fresh gonad weights were not available due to the necessity of freezing all samples before processing.

Sampling In all, 2014 boarfish were collected from January to December 2010, excluding August (Table 1). Samples were collected from ICES Areas VIa, VIIb, VIIc, VIIg, VIIh, VIIj, VIIIa, and VIIId from both commercial and research vessels. Due to the industrial nature of the fishery (RSW tank vessels), high-quality fresh samples could not be collected and processed on board; therefore, all samples were frozen on board before being returned to the laboratory for processing (Figure 1). Samples were fully thawed before processing. Each fish was measured for total length (LT) to the 0.5 cm below and total body weight (WT) to the nearest 0.01 g. The body cavity was opened with a semi-circle incision running anteriorly from the cloaca to the pectoral fin and sex was assessed. The organs were dissected out and carefully separated. Gutted weight (WG, g) was recorded. Ovaries and testes

Oocyte development and maturity classification To examine oocyte development and microscopically verify the precision of the macroscopic maturity scale, a subsample of 164 ovaries and 21 testes were fixed in 10% buffered formalin for histological analyses (Table 2). Once fixed, a 4-mm thick transverse section was taken from the middle of each gonad, dehydrated through a series of alcohol and solvent solutions and infiltrated with paraffin wax on an automatic tissue processor (VIPw, Tissue-Tekw, the Netherlands). A rotary microtome (HM 325, Thermo Fischer Scientific Inc., Germany) was used to cut 4-mm thick sections, which were stained with haematoxylin and eosin, cover-slipped with a mounting medium, and examined under a Leica DM 2000 light microscope with a Leica DFC490 digital camera.


Sex (LT cm) Female 6.5–7.0 7.5–8.0 8.5–9.0 9.5–10.0 10.5 –11.0 11.5 –12.0 12.5 –13.0 13.5 –14.0 14.5 –15.0 15.5 –16.0 16.5 –17.0 17.5 –18.0 Total Male 5.5–6.0 6.5–7.0 7.5–8.0 8.5–9.0 9.5–10.0 10.5 –11.0 11.5 –12.0 12.5 –13.0 13.5 –14.0 14.5 –15.0 15.5 –16.0 Total Unsexed 2.5–3.0 3.5–4.0 4.5–5.0 5.5–6.0 6.5–7.0 7.5–8.0 8.5–9.0 9.5–10.0 Total

500

E. D. Farrell et al.


Figure 1. The distribution of C. aper samples collected in the Northeast Atlantic. Thirteen of the highest quality sectioned ovaries were selected for measuring oocyte developmental stages (Table 2). Leica Application Suite (2.6.0 R1) was used to take images of the sectioned ovaries and to measure oocyte diameters. Only oocytes sectioned through the nucleus (n ¼ 190) were measured. Due to the necessity of using previously frozen samples, histological processing, and natural variation, oocytes were not perfectly spherical in shape; therefore, maximal and minimal oocyte diameters were averaged to decrease variance and avoid artificially increasing the overlap between different groups of

oocytes (West, 1990). Hydrated oocytes rarely survived histological processing intact and were observed to be highly crenulated, which made accurate measurements difficult. Therefore, additional measurements were made of the largest oocytes extracted directly from the ovaries of spawning females, preserved in 10% buffered formalin, and rinsed in deionised water before taking measurements. Five ripe ovaries (May n ¼ 3, June n ¼ 2) were assessed for oocyte size frequency distribution. Between 280 and 496, oocytes, sectioned through the nucleus, were measured from

Oocyte development and maturity classification of boarfish

501

Table 2. Capros aper specimens per sex, length range, and maturity stage used in histological analyses.

each ovary. In total, 1872 oocytes were measured. The percentage frequency was plotted against oocyte diameter. All 185 sectioned gonads were maturity staged. Testes were primarily examined for the presence or the absence of sperm, while ovaries were classified by the most advanced type of oocyte present. These microscopic observations were combined with macroscopic morphological observations of ovaries and testes and the gonadosomatic index (GSI) values to develop a comprehensive, eight-point maturity scale (Table 3). The maturity of all 2014 boarfish was assessed based on the macroscopic appearance of the gonads using this scale.

2

3

4

5

6

7

8

Total

1 1 6 1 – – – – – – – – 9

– – 1 (1) 2 – 3 (1) 5 3 12 (3) 2 (1) 1 – 29

– – – – 1 2 1 3 3 2 – – 12

– – – – 1 2 (1) 4 5 (1) 4 (1) 7 (1) – – 23

– – – – 1 2 (1) 5 2 1 4 (2) – – 15

– – – 3 6 9 10 13 10 1 – 1 53

– – – – 1 2 4 5 6 1 3 1 23

1 1 7 6 10 20 29 31 36 17 4 2 164

1 – – – – – – – 1

– – 1 1 1 1 3 2 9

– – – – 1 3

– – – – – – – – 0

– – – – – – – 1 1

– – – – – 1 – 1 2

– – – – – 1 1 – 2

1 0 1 1 2 6 4 6 21

2 6

Length-at-maturity and reproductive cycle Maturity stages 1 and 2 were considered immature and maturity stages 3 –8 were considered mature. The length at which 50% of males and females were sexually mature (L50) was calculated by the logistic regression of binomial maturity data (immature ¼ 0, mature ¼ 1) of the 2014 boarfish. The proportional monthly occurrence of maturity stages was plotted for mature males and females to investigate the reproductive cycle. The GSI was calculated and plotted against LT (all samples) and month (mature samples only) to determine maturity and seasonal reproductive patterns, respectively. GSI =

Values in parentheses are the 13 specimens measured for oocyte diameter.

100 WOT . WG

(1)

Table 3. Macroscopic and histological description of the maturity stages of C. aper. Maturity stage 1. Virgin 2. Developing virgin 3. Early maturing

4. Mature

5. Ripe

6. Spawning

7. Spent/ resorption

8. Resting

Female Tiny, thread-like, and transparent (WOT , 0.01 g and GSI ≤ 0.08). Cannot distinguish sex Ovaries very small (WOT ≤ 0.20 g and GSI ≤ 1) and opaque. Ovary occupies 5% of body cavity. All oocytes are in the primary growth phase Ovaries developed but small and pale white colour (WOT ¼ 0.2–2.5 g and GSI ¼ 1.0–3.5). Firm to the touch. Secondary growth phase initiated with the formation of cortical alveoli. Oocytes in primary growth phase also present Ovaries not as swollen as stage 5, occupying 30 –40% body cavity (WOT ¼ 0.5–3.9 g and GSI ¼ 1.2–5.5). No hydrated eggs present. More pink in colour than stage 3 but may also be orange. Most advanced oocytes undergoing vitellogenesis but previous stages also present Ovaries heavily vascularised and obviously swollen, occupying 40–50% of body cavity (WOT ¼ 1.0–6.0 g and GSI ¼ 2.4– 10.0). Hydrated eggs present. Colour varies from whitish to deep orange. Ovaries do not run under pressure. All previous oocyte stages present Ovaries large and filling over 50% of body cavity (WOT ¼ 1.0 – 8.0 g and GSI ¼ 3.7– 13.4). Lots of hydrated eggs clearly visible. Ovary membrane thin and easily torn. Runs under pressure. All previous oocyte stages present The ovary is still quite large, 30% of body cavity (WOT ¼ 0.6– 3.0 g and GSI ¼ 2.0– 6.0). Look similar to stage 4 but may have a bruised appearance. Generalized atresia of mature oocytes. Ovary appears to have a denser structure Ovaries reduced in size and white (WOT ¼ 0.2–2.5 g and GSI ¼ 1– 3). Similar to stage 3, except that they are softer to the touch. Remnants of atretic oocytes still present and lamellae are more loosely organized than stage 3

Values for gonad weight and GSI are a guideline only.

Male Tiny, thread-like, and transparent (WOT , 0.01 g and GSI ≤ 0.08). Cannot distinguish sex Testes small (WOT ≤ 0.1 g and GSI ¼ 0.05 –0.7) and opaque. Still slightly round in appearance, beginning to develop characteristic leaf-like lobes. Small amount of sperm present Testes small but obviously developed (WOT ¼ 0.03 –0.76 g and GSI ¼ 0.14 –1.60). Pale white in colour

Testes nearly full size but do not appear swollen or obviously white (WOT ¼ 0.15 –2.9 g and GSI ¼ 0.47 –4.75). No sperm visible when cut

Testes very large, white, and swollen (WOT ¼ 0.25 –2.9 g and GSI ¼ 1.0–5.0). Sperm visible when cut

Testes are the same as stage 5 except that sperm runs under pressure (WOT ¼ 0.5 –2.2 g and GSI ¼ 1.6 –4.0) Testes appear bruised and reduced in size (WOT ¼ 0.1 –1.4 g and GSI ¼ 0.5–2.5). Some small amounts of sperm may remain Testes small, similar size as stage 3 (WOT ¼ 0.03 –0.70 g and GSI ¼ 0.13 –1.40). Pinkish in colour


Maturity Stage Female 6.5–7.0 7.5–8.0 8.5–9.0 9.5–10.0 10.5 –11 11.5 –12.0 12.5 –13.0 13.5 –14.0 14.5 –15.0 15.5 –16.0 16.5 –17.0 17.5 –18.0 Total Male 7.5–8.0 8.5–9.0 9.5–10.0 10.5 –11.0 11.5 –12.0 12.5 –13.0 13.5 –14.0 14.5 –15.0 Total

502


Table 4. The stages of oocyte development of C. aper and average oocyte and nuclear diameters. Stage Chromatin nucleolus Early perinucleolus Late perinucleolus Cortical alveolus Vitellogenesis Maturation Hydration Whole mount hydrated

Oocyte stage 1a 1b 1c 2a 2b 3a 3b –

Number measured 8 38 26 21 37 41 9 10

Average oocyte diameter (mm) 30 (23 –40) 61 (34 –93) 115 (74 –151) 178 (144– 216) 293 (213– 455) 470 (375– 566) 758 (630– 951) 1 002 (985– 1 021)

Average nucleus diameter (mm) 13 (11 –16) 30 (17 –44) 51 (35 –66) 70 (49 –92) 77 (35 –121) 80 (51 –146) – Oil drop 197 (188 –207)

The range of diameters is shown in parentheses.

HSI =

100 WL . WG

(2)

Results Sampling Samples comprised 1066 females, 833 males, and 115 unsexed immature individuals (Table 1). Mature male and female boarfish were sexually dimorphic and could be readily identified based on external characteristics. Mature males had a red and white pattern of vertical bands on their lateral flanks, which become more pronounced during spawning. Mature females lacked these vertical stripes, were a uniform pale red/orange colour, and reached a greater maximum LT than males.

Oocyte development and maturity classification The process of oocyte development in boarfish followed the same basic progression as that described for other small pelagic species such as the slender snipefish, Macrorhamphosus gracilis (Arruda, 1988). Seven separate oocyte developmental stages were identified; however, it should be noted that development is a continuum and transitional stages may account for the overlap in oocyte diameters (Table 4, Figure 2). Oogonia were observed to undergo a short period of primary growth to enter the chromatin nucleolus stage (oocyte stage 1a), which were small round cells with a thin basophilic cytoplasm (Table 4, Figure 2a). Early perinucleolar oocytes (oocyte stage 1b) had a large nucleus containing several nucleoli, occupying the peripheral part of the nucleus. As the oocytes enlarged, the amount of strongly basophilic cytoplasm increased (Table 4, Figure 2b). During the late perinucleolus stage (oocyte stage 1c), the chromatin material was dispersed throughout the nucleus, causing the nucleus to appear granular. The cytoplasm was divided into two concentric zones: the inner zone was basophilic and the outer follicular layer was less densely stained (Table 4, Figure 2c). The secondary growth phase was initiated by the appearance of the cortical alveoli (oocyte stage 2a). The cytoplasm lost some of its basophilic property and a narrow zone of cortical alveoli formed in the periphery of the cytoplasm. As the oocytes increased in size so did the number and size of the cortical alveoli. The zona radiata began to develop towards the end of this stage. Vitellogenesis (oocyte stage 2b) was characterized by the appearance of yolk granules (Table 4, Figure 2e). During

maturation (oocyte stage 3a), the yolk granules increased in size and fused to form a continuous mass of yolk (Table 4, Figure 2f). The lipid droplets also coalesced just before the nucleus migrating to the periphery of the oocyte and the nuclear membrane dissolving. The final stage (oocyte stage 3b) involved the rapid increase in diameter of the oocytes through hydration (Table 4, Figure 2g and h). Oocytes of all developmental stages were observed in ripe and spawning ovaries. No post-ovulatory follicles were observed. The ovaries of the five stage 5 (ripe) females also contained oocytes at all stages of development (Figure 3), indicating asynchronous oocyte development. Pre-vitellogenic oocytes (oocyte stages 1a– 2a) accounted for 87% of the measured oocytes, while vitellogenic oocytes (oocyte stages 2b– 3b) accounted for just 13%. There was no apparent hiatus between pre-vitellogenic and vitellogenic oocytes (Figure 3). Due to histological processing the largest hydrated oocytes were damaged and could not be measured accurately. The ovaries and testes of immature (maturity stage 1) specimens were tiny, transparent, and thread-like and could not be distinguished or histologically processed. These specimens were recorded as unsexed. The ovaries of developing virgins (maturity stage 2) comprised oocytes in the primary growth phase only (oocyte stages 1a, b, and c). The ovaries were well structured with clearly defined lamellae and had no evidence of atresia (Figure 4a). The oocytes of early maturing ovaries (maturity stage 3) had initiated the secondary growth phase with the formation of cortical alveoli (oocyte stage 2a; Figure 4b). In mature ovaries (maturity stage 4), the most advanced oocytes were undergoing vitellogenesis; however, no hydrated oocytes were present (Figure 4c). Ripe and spawning ovaries (maturity stages 5 and 6) had a looser structure and an abundance of maturing and hydrated oocytes (oocyte stages 3a and 3b; Figure 4d). There were no obvious histological differences observed between these stages. The only difference was observed in whole, spawning fish; the oocytes of which were easily extruded with slight pressure on the abdomen. Ripe ovaries also still had a large proportion of primary and secondary oocytes. The ovaries of spent fish (Maturity stage 7) underwent generalized atresia, which occurred in two phases. Hydrated oocytes were resorbed first, followed by vitellogenic oocytes (Figure 4e). As vitellogenic oocytes were resorbed, the structure of the ovary became denser. Resting ovaries (maturity stage 8) still had a high abundance of the remnants of resorbed oocytes and a loose lamellar structure, which was reformed by the time resting ovaries returned to the early maturing stage. No abnormal maturity stages were observed during the current study and as such an abnormal stage was not included in the maturity scale. However, when routine sampling of boarfish


The liver weight (WL) of each boarfish was measured to the nearest 0.01 g. The hepatosomatic index (HSI) was calculated as a measure of condition and plotted against LT (all samples) and month (mature samples only) to determine length-related and seasonal patterns, respectively.

503



Figure 2. Stages of oocyte development in C. aper. (a) Chromatin nucleolus stage, (b) early perinucleolus stage, (c) late perinucleolus stage, (d) cortical alveolus stage, (e) vitellogenesis, (f) oocyte maturation, (g) hydration, and (h) whole mount hydrated oocyte. Scales bars equal 50 mm.

is established, it should be noted in the sampling protocol that abnormal stages be recorded.

Length-at-maturity and reproductive cycle The average GSI (Figure 5a) displayed a steep increase for both males and females from 8.5 –9.5 cm LT onwards. The average female GSI appeared to plateau at 11.5 cm LT, whereas the average male GSI appeared to continue on an upward trend. The average HSI (Figure 5b) also displayed an increasing trend for males and females up to 11 cm LT before declining. The male L50 was 9.7 cm LT with upper and lower 95% CI of 9.6 and 9.8 cm LT [Equation (3), Figure 6]. The largest immature male

was 10.5 cm LT and the smallest mature was 9.0 cm LT. The female L50 was 9.7 cm LT with upper and lower 95% CI of 9.6 and 9.9 cm LT [Equation (4), Figure 6]. The largest immature female was 11.0 cm LT and the smallest mature was 9.0 cm LT. P(LT ) =

1 , 1 + e33.36+(−3.44) LT

(3)

P(LT ) =

1 , 1 + e31.89+(−3.29)LT

(4)

where P(LT) is the proportion mature at length (LT).

504 The monthly distribution of maturity stages indicated an annual reproductive pattern for both males and females (Figure 7). A small number of stage 3 females (n ¼ 7) was observed

in December. These females ranged from 9.5 –10.0 cm LT and were likely maturing (entering stage 3) for the first time. For previously mature males and females, the reproductive cycle began as early as February with the appearance of stage 3 (early maturing) individuals. By April, all specimens were either stage 3 or 4 (mature). A small proportion of stage 5 (ripe) fish were observed in May. Both males and females were observed to spawn in June and July, though the proportion of spawning ovaries was higher in July. No samples were available for August. Spawning had ceased by September and the majority of females were in stage 7 (spent/resorption), whereas the majority of males sampled had advanced to stage 8 (Resting). Stage 8 was the dominant maturity stage from November to January. The GSI by month for mature males and females followed the same trends as the occurrence of the maturity stages (Figure 8a). The GSI increased from April onwards, with a significant peak in July before decreasing from September to the end of the year. The female GSI was proportionately higher than the male GSI. The HSI by month for mature males and females also followed the same trends as the occurrence of the maturity stages and the GSI (Figure 8b). HSI increased from April to a peak in July before decreasing from September to the end of the year.

Figure 4. Transverse sections of C. aper ovaries. (a) Stage 2, developing virgin; (b) stage 3, early maturing; (c) stage 4, mature; (d) stages 5 and 6, ripe and spawning; (e) stage 7, spent resorption; and (f) stage 8, resting. Scale bars equal 200 mm.


Figure 3. The oocyte size frequency distribution in stage 5 (ripe) ovaries (n ¼ 5) of C. aper.



Figure 7. The occurrence of the maturity stages of mature C. aper by month for (a) males and (b) females. The number under each month indicates the number of samples.

Figure 6. Maturity ogives for total length (LT) for C. aper. The L50 for males and females was 9.7 cm LT.

Discussion Accurate and objective maturity determination is essential for the assessment and management of exploited stocks (Vitale et al., 2006). Although maturity staging based on the external appearance of the gonads is the simplest, most rapid, and widely applied method, it should be validated by histology to ensure accuracy (West, 1990). The combined use of macroscopic morphological observations, histological analyses, and biological indices in the current study presents the most comprehensive investigation of oocyte development and maturity classification of C. aper to date. Previous studies relied solely on macroscopic observations

Figure 8. The average (a) GSI and (b) HSI of mature (stages 3 – 8) male and female C. aper per month. The error bars indicate standard deviation.


Figure 5. The average (a) GSI and (b) HSI by total length for C. aper.

505

506

boarfish in the aquarium were observed to spawn regularly every 2 –3 d for a period of 9 months, whereas males spawned daily (J. Hemdal, Toledo Zoo Aquarium, pers. comm.). These observations provide evidence that individual boarfish are physiologically capable of spawning repeatedly over an extended period. The primary focus of the current study was to establish important baseline data on the oocyte development and maturity classification from which further studies may develop. Therefore, the examination of reproductive strategy and fecundity modality can be viewed as preliminary findings on which future specific studies can elaborate. Verification of reproductive strategy and fecundity modality is an inherently difficult task as corroborated by the multitude of studies on species such as horse mackerel (Trachurus trachurus L.) and hake (Merluccius merluccius L.; Murua and Motos, 2006; Gordo et al., 2008; Ndjaula et al., 2009). There are four generally accepted lines of evidence for distinguishing determinate and indeterminate fecundity (Murua et al., 1998). A distinct hiatus between pre-vitellogenic and vitellogenic oocytes indicates that fecundity is determinate and a lack of a hiatus may indicate that fecundity is indeterminate. A seasonal decrease in the stock of vitellogenic oocytes and also an increase in their mean diameter indicate determinate fecundity. Finally, fish with indeterminate fecundity exhibit generalized atresia and resorption of oocytes at the end of the spawning season (West, 1990). The preliminary data in the current study demonstrated that boarfish appear to fulfil at least two of these criteria; lack of a hiatus between pre-vitellogenic and vitellogenic oocytes and a generalized atresia and resorption of mature oocytes at the end of the spawning season. Coupled with the asynchronous oocyte development, the increase in HSI during the spawning season, and the extended spawning period observed in captive boarfish, all indications are that boarfish have an indeterminate fecundity. However, a further specific investigation of fecundity modality with an increased sample size is required to confirm this. Given that all available evidence indicates that boarfish are capable of spawning regularly for an extended period, potential annual fecundity is likely to be significant and highly variable. It will largely be determined by the physical and biological conditions in a given year (Hunter and Leong, 1981). The current study indicates that spawning in the Northeast Atlantic in 2010 was restricted to the summer with a significant peak in July. It may be beneficial to identify the environmental conditions in the known spawning areas during this period and analyse those of previous spawning seasons. Such an exercise may go some way towards explaining the increase in abundance and distribution of boarfish. Blanchard and Vandermeirsch (2005) attributed this increase to an increase in water temperature, which resulted in enhanced recruitment. However, current evidence suggests that prey availability may also have a significant role and specifically the abundance of the primary prey item, the copepod Calanus helgolandicus (Lopes et al., 2006). Further analyses of the distribution and abundance of C. helgolandicus during the periods of boarfish population “blooms” may also provide useful indicators for future abundance trends. While the boarfish fishery is a relatively recent phenomenon, a sound basis for the development of management advice and for future research has now been established and further studies on age and growth characteristics are already underway. This base knowledge will enable routine sampling and further more specific biological studies to be initiated. These studies should perhaps prioritize detailed investigation of fecundity modality and also the


¨ zaydin, 1996; without histology or biological indices (Kaya and O White et al., 2011). The current study goes further to establish and verify a species-specific maturity scale. This is of increasing importance as the fishery develops and fisheries independent surveys and routine sampling are established. The reported length-at-maturity of 9.7 cm LT for male and female boarfish was supported by the marked changes in both ¨ zaydin (1996) the GSI and HSI around this length. Kaya and O reported length- and age-at-maturity of boarfish in Turkish waters to be 8.5 cm and 2 years. However, the lack of information on the method of maturity estimation in this study precludes meaningful comparisons. White et al. (2011) estimated male and female length-at-maturity in the Northeast Atlantic, based on standard length (LS), to occur at 8.8 and 8.1 cm LS, corresponding to 11.0 and 10.1 cm LT, respectively (LT ¼ 1.2113 LS + 3.7227, E. D. Farrell, unpublished data). These values are higher than the current values, perhaps due to the relatively small sample size used. White et al. (2011) also estimated age-at-maturity to be 5.25 years, which may be an overestimate. Although the age of the specimens in the current study is not yet established, a recent study of marginal increment analyses of boarfish otoliths indicated rapid growth in the first 2 –3 years, after which growth levelled off (Hu¨ssy et al., 2011). This plateau in growth may be related to the attainment of maturity as energy is allocated to other processes such as reproduction, although a more detailed investigation of age and growth is required. The patterns observed in both the GSI and HSI with LT also lend support to this theory. The sharp increase in GSI with LT observed at maturity and the corresponding decrease in HSI with LT may be indicative of the energy resources, which are redeployed from growth to reproduction. Boarfish appear to be similar to many other small pelagic shoaling species (Petitgas, 2010), in that maturity occurs at a small LT after a short initial period of rapid growth (Hu¨ssy et al., 2011). The reproductive cycle of boarfish appears to be well defined, and both the monthly distribution of maturity stages and the monthly GSI and HSI indicate that it commences between February and April with the appearance of stage 3 early maturing fish and finishes between October and December when fish enter the resting stage. This annual cycle was also identified in the growth pattern of boarfish otoliths (Hu¨ssy et al., 2011). Otolith growth was fastest from March/April to September/October, during the reproductive cycle, before slowing significantly during winter. A decrease in the monthly HSI of mature fish might be expected during the spawning season, as energy is redeployed into gonad maturation. However, the converse was observed with a HSI peak in June and July for both males and females. In females, this may be related to the observed asynchronous oocyte development. This mode of oocyte development requires a substantial and continuous energy investment to sustain it (Hunter and Leong, 1981); therefore, the increase in HSI may be an indication of increased feeding activity during this time. This strategy, where spawning depends on the continued intake of energy from feeding during the spawning season, is known as income spawning (Armstrong and Witthames, 2012). It is an adaptation that enables small bodied fish to have a relatively high fecundity per unit body mass and also to increase their annual fecundity when feeding conditions are favourable (Armstrong and Witthames, 2012). Observations of captive boarfish in Toledo Zoo Aquarium have shown that they are capable of spawning for an extended period provided suitable conditions are maintained. Three female


Oocyte development and maturity classification of boarfish assessment of the error associated with macroscopic maturity observations during routine sampling.

Acknowledgements This work was funded by the Killybegs Fishermen’s Organisation, Killybegs, Co. Donegal, Ireland, and E. D. Farrell would like to thank all the skippers and crew of the Killybegs pelagic fleet for helping with sample collection. We are also grateful to all the scientists and crew of the RVs “Celtic Explorer”, “Scotia”, “Tridens”, “Walther Herwig III”, “Johan Hjort”, and “Thalassa” for helping with the collection of samples.

References

Hunter, J. R., and Leong, R. 1981. The spawning energetics of female northern anchovy, Engraulis mordax. Fishery Bulletin US, 79: 215– 230. Hunter, J. R., Maciewicz, B. J., Lo, N. C. H., and Kimbrell, C. A. 1992. Fecundity, spawning, and maturity of female Dover sole Microstomus pacificus, with an evaluation of assumptions and precision. Fishery Bulletin US, 90: 101– 128. Hu¨ssy, K., Coad, J. O., Farrell, E. D., Clausen, L. W., and Clarke, M. W. 2012. Age verification of boarfish (Capros aper) in the Northeast Atlantic Ocean. ICES Journal of Marine Science, 69: 34 – 40. ¨ zaydin, O. 1996. A preliminary investigation on Kaya, M., and O biology of Capros aper (L.1758) (Pisces: Caproidae). Turkish Journal of Zoology, 20: 51– 55. Lopes, M., Murta, A. G., and Cabral, H. N. 2006. The ecological significance of the zooplanktivores, snipefish Macroramphosus spp. and boarfish Capros aper, in the food web of the south-east North Atlantic. Journal of Fish Biology, 69: 363 – 378. Morgan, M. J. 2008. Integrating reproductive biology into scientific advice for fisheries management. Journal of Northwest Atlantic Fishery Science, 41: 37 – 51. Murua, H., and Motos, L. 2006. Reproductive strategy and spawning activity of the European hake Merluccius merluccius (L.) in the Bay of Biscay. Journal of Fish Biology, 69: 1288 – 1303. Murua, H., Motos, L., and Lucio, P. 1998. Reproductive modality and batch fecundity of the European hake (Merluccius merluccius L.) in the Bay of Biscay. CalCOFI Reports, 39: 196 – 203. Murua, H., and Saborido-Rey, F. 2003. Female reproductive strategies of marine fish species of the North Atlantic. Journal of Northwest Atlantic Fishery Science, 33: 23– 31. Ndjaula, H. O. N., Hansen, T., Kruger-Johnsen, M., and Kjesbu, O. S. 2009. Oocyte development in captive Atlantic horse mackerel Trachurus trachurus. ICES Journal of Marine Science, 66: 623 – 630. Petitgas, P. (Ed.) 2010. Life cycle spatial patterns of small pelagic fish in the Northeast Atlantic. ICES Cooperative Research Report No. 306. 93 pp. http://www.ices.dk/pubs/crr/crr306/CRR%20306Web.pdf. Que´ro, J. C. 1986. Caproidae. In Fishes of the North-eastern Atlantic and the Mediterranean, pp. 777 –779. Ed. by P. J. P. Whitehead, M. L. Bauchot, J. C. Hureau, J. Nielsen, and E. Tortonese. UNESCO, Paris. Vitale, F., Sveda¨ng, H., and Cardinale, M. 2006. Histological analysis invalidates macroscopically determined maturity ogives of the Kattegat cod (Gadus morhua) and suggests new proxies for estimating maturity status of individual fish. ICES Journal of Marine Science, 63: 485– 492. West, G. J. 1990. Methods of assessing ovarian development in fishes: a review. Australian Journal of Marine and Freshwater Research, 41: 199– 222. White, E., Minto, C., Nolan, C. P., King, E., Mullins, E., and Clarke, M. W. 2011. First estimates of age, growth, and maturity of boarfish (Capros aper): a species newly exploited in the Northeast Atlantic. ICES Journal of Marine Science, 68: 61 – 66.

Handling editor: Rochelle Seitz


Armstrong, M. J., and Witthames, P. R. 2012. Developments in understanding of fecundity of fish stocks in relation to egg production methods for estimating spawning stock biomass. Fisheries Research, 117-118: 35– 47. Arruda, L. M. 1988. Maturation cycle in the female gonad of the snipefish, Macrorhamphosus gracilis (Lowe, 1839) (Gasterosteiformes, Macrorhamphosidae), off the western coast of Portugal. Investigacioń Pesquera, 52: 355 –374. Blanchard, F., and Vandermeirsch, F. 2005. Warming and exponential abundance increase of the subtropical fish Capros aper in the Bay of Biscay (1973– 2002). Comptes Rendus Biologies, 328: 505 – 509. Borges, L., van Keeken, O. A., van Helmond, A. T. M., Couperus, B., and Dickey-Collas, M. 2008. What do pelagic freezer-trawlers discard? ICES Journal of Marine Science, 65: 605 – 611. Cunningham, J. T. 1888. Notes and memoranda. Some notes on Plymouth fishes. Journal of the Marine Biological Association of the UK, 2: 234 – 250. Cunningham, J. T. 1889. Studies of the reproduction and development of teleostean fishes occurring in the neighbourhood of Plymouth. Journal of the Marine Biological Association of the UK, 1: 10– 54. Ehrenbaum, E. 1905. Eier und larven von fischen. In Nordisches Plankton. Lipsius & Tischer, Kiel. Fonseca, P., Campos, A., Larsen, R. B., Borges, T. C., and Erzini, K. 2005. Using a modified Nørdmore grid for by-catch reduction in the Portuguese crustacean-trawl fishery. Fisheries Research, 71: 223– 239. Gordo, L. S., Costa, A., Abaunza, P., Lucio, P., Eltink, A., and Figueiredo, I. 2008. Determinate versus indeterminate fecundity in horse mackerel. Fisheries Research, 89: 181 – 185. Hefford, A. E. 1910. Notes on the teleostean ova and larvae observed at Plymouth in spring and summer, 1909. Journal of the Marine Biological Association of the UK, 9: 1 – 58. Holgersen, H. 1954. First record of the boarfish, Capros aper in Norway. Aktietrykkeriet I Stavanger, Stavanger Museum, 17: 1 –11. Holt, E. W. L., and Scott, S. D. 1898. A Record of the Teleostean Eggs and Larvæ observed at Plymouth in 1897. Journal of the Marine Biological Association of the UK, 5: 156 –171.

507