Marine Biology (1999) 135: 315±319
Ó Springer-Verlag 1999
W. H. H. Sauer á Y. C. Melo á W. de Wet
Fecundity of the chokka squid Loligo vulgaris reynaudii on the southeastern coast of South Africa
Received: 9 October 1998 / Accepted: 22 April 1999
Abstract Potential fecundity in Loligo vulgaris reynaudii was estimated to be about 17 000 eggs, calculated as the total number of discernible oocytes in the ovary and oviduct. Squid were observed to spawn up to 8140 eggs over a 36 h period in captivity. First estimations of actual fecundity are therefore between 8000 and 17 000 eggs. Factors complicating a more accurate estimation of actual fecundity in this species include diculties with aquarium maintenance, their behaviour of spawning over a protracted period and in multiple sites, and atretic oocytes observed in both developing and mature ovaries. Detailed morphological and histological analysis of gonads collected at regular intervals over a complete spawning season will allow a more precise calculation of actual fecundity.
Introduction Studies on the reproduction and population biology of several squid species of the genus Loligo have been made worldwide (e.g. Holme 1974; Martins 1982; Hixon 1983; Hat®eld et al. 1990; Lum-Kong et al. 1992). These and many other studies have established a relatively high degree of conformity in the breeding biology of loliginid
Communicated by O. Kinne, Oldendorf/Luhe W.H.H. Sauer (&) Department of Ichthyology and Fisheries Science, P.O. Box 94, Rhodes University, Grahamstown, 6140, Republic of South Africa Y.C. Melo Sea Fisheries Research Institute, Private Bag X2, Roggebaai, 8012, Cape Town, Republic of South Africa W. de Wet Port Elizabeth Museum, P.O. Box 13147, Humewood, 6013, Port Elizabeth, Republic of South Africa
squid. De®ned geographical spawning areas, patterns of migration within coastal waters, an annual life cycle and an assumption of semelparity or terminal spawning, are all loliginid characteristics (Boyle et al. 1995). Many mysteries still remain, for example, the origin of multiple recruitment modes, the interpretation of the extended breeding season in some species (whether or not this is a characteristic of the individual or the population), the time course of individual spawnings and whether these are batch spawnings (Harman et al. 1989; Lewis and Choat 1993; Boyle et al. 1995; Moltschaniwskyj 1995). Determination of actual fecundity in loliginid squid is not simply a count of viable eggs in the ovary and/or oviduct. For an accurate determination of fecundity answers to the above questions are crucial. Although studies on the reproduction and population biology of Loligo vulgaris reynaudii have been ongoing since the early 1980s, and a review by Augustyn et al. (1994) shows a fairly good understanding of the life history of this species, a lack of knowledge on maturation and fecundity is apparent. Badenhorst (1974), Sauer and Lipinski (1990) and Lipinski and Underhill (1995) conducted studies on maturation. However, those studies were based on histological and macroscopic analysis of the gonads. To date, no study has been made on the fecundity of L. v. reynaudii. Estimation of egg production for Loligo vulgaris reynaudii is complicated by their serial spawning behaviour. Chokka squid (Melo and Sauer 1999) deposit egg strands over an extended period within one season. Furthermore, recent histological analysis of the gonads of squid of various maturity stages and throughout their distribution range shows that ovarian atresia is present in both developing and mature ovaries (Melo and Sauer 1998), and must be taken into account. This paper explores the conventional methods of estimating egg production in loliginid squid, provides a ®rst estimate of actual fecundity for L. v. reynaudii and suggests a future research approach that may provide a more accurate determination.
316
Materials and methods The inshore spawning grounds and the distribution range of Loligo vulgaris reynaudii on the southeastern coast of South Africa are shown in Fig. 1 of Melo and Sauer (1999, this issue). Variably mature squid (Stage 3 and Stage 5, Lipinski 1979) were obtained from demersal trawls carried out at depths between 48 and 117 m during September 1992 and May 1993. A further sample of mature squid was obtained on the inshore spawning grounds (<50 m) by hand jigging. It was also possible to obtain partially spent squid on the inshore spawning grounds with SCUBA and a Hawaiian sling, after observing an individual depositing an egg strand. Squid were weighed to the nearest gram, and the dorsal mantle length measured to the nearest millimeter. The entire gonad, including ovaries and oviducal complexes in mature individuals, were removed and initially preserved in Bouin's solution and thereafter in 50% alcohol. After removal from the alcohol, the ovary and oviduct were separated in the laboratory, dried with a paper towel and weighed on a Sartorius 2463 scale (GoÈttingen) to the nearest milligram. Three subsamples of 0.25 g were removed from both the ovary and oviduct and stored in 50% alcohol, and the eggs counted under a dissection microscope at ´20 magni®cation. Potential fecundity was determined as follows. For each subsample three counts were done of: (i) the total number of oocytes distinguishable in the ovary (FecOv), (ii) the number of oocytes greater than Stage 4 (Cowden 1968) in the ovary (FecOv4) (referred to as advanced oocytes in the text), and (iii) the total number of mature eggs in the oviduct (FecOd). The total number of eggs (TN) was then estimated using the relationship: X X TN Wt = Wi Ni where Wt is the total weight of eggs, Wi is the weight of oocytes/ eggs in sample i from the ovary or oviduct, and Ni is the number of oocytes/eggs in sample i from the ovary or oviduct. Gonad indices (GSI) were de®ned as the total gonad weight divided by the body weight. ScheeÂ's multiple comparison procedure was used to compare variables, as sample sizes varied. To determine actual fecundity, pairs of live squid (male and female) were placed in a 15 m diameter, 2 m deep tank, and egg strands were collected after spawning had ceased. Only one female was present in the tank at any one time. The number of eggs spawned was calculated from an average number of eggs per strand (Sauer et al. 1993) multiplied by the number of strands deposited. In all cases, the females died within a few days after ®rst spawning. Where possible, dead squid were removed from the tank and a count made of the remaining eggs in the ovary and oviduct.
Results Tables 1 and 2 summarize and compare the estimates of potential fecundity. The results are separated into three groups namely: (a) Stage 5 individuals (mature) collected oshore, (b) mature squid jigged on the inshore spawning grounds, and (c) partially spent squid (after a diver had observed the deposition of an egg strand). A signi®cant increase (p < 0.05) was found in both the total and advanced oocyte numbers for squid obtained inshore (Table 2), although the number of eggs in the oviducts was signi®cantly less (p < 0.01) than in those collected oshore. A particularly interesting result was the fact that on the inshore spawning grounds jigged squid were found to have signi®cantly lower (p < 0.01) numbers of eggs in the oviducts than those collected after observing an
egg strand deposited (partially spent squid, Table 2). It is possible that squid go through a resting phase (minutes or hours) between successive spawning bouts during which time the oviduct re®lls with eggs. During this time they may be more susceptible to the jigs. A realistic estimate of potential fecundity was therefore calculated as 17 809 eggs, the mean value of the number of discernible oocytes in the ovary of squid jigged on the spawning grounds added to the number of eggs in the oviducts of partially spent squid (Table 1). The results of the GSI calculations show the sensitivity of this method to the number of eggs in the oviTable 1 Loligo vulgaris reynaudii. Summary of potential fecundity data [ML mantle length (mm); WT body weight (g); Ovary ovary weight (g); Oviduc oviduct weight (g); FecOv number of all the oocytes in the ovary; FecOv4 number of oocytes greater than Stage 4 (Cowden 1968) in the ovary; FecOd number of mature eggs in the oviduct; GSI gonadosomatic indices] Variable
Mean
SD
Min.
Max.
19.75 46.67 2.52 3.36 4708 2519 1198 1.74
133.00 71.00 1.66 0.33 5692 2297 104 2.81
222.00 295.00 14.42 16.68 28691 14728 6034 10.59
Spawning ground (jigged) N = 25 ML 190.96 23.58 WT 174.40 61.78 Ovary 7.19 2.07 Oviduc 1.64 2.29 FecOv 16472 3708 FecOv4 8075 3118 FecOd 654 1033 GSI 5.32 1.72
144.00 75.00 3.33 0.09 8894 3754 6 3.17
233.00 285.00 10.58 9.22 24118 11968 4056 9.88
Partially spent squid N = 25 ML 192.08 14.86 WT 194.60 51.74 Ovary 6.10 2.29 Oviduc 3.60 3.17 FecOv 15611 3688 FecOv4 6563 1307 FecOd 1337 1225 GSI 4.90 1.58
163.00 105.00 2.80 0.03 7942 4030 5 2.70
215.00 285.00 12.39 11.43 25334 9419 4028 7.46
Oshore N = 76 ML 174.96 WT 158.72 Ovary 6.85 Oviduc 3.66 FecOv 13 853 FecOv4 6804 FecOd 1342 GSI 6.45
Table 2 Loligo vulgaris reynaudii. Results of ScheeÂ's multiple comparison procedure. Abbreviations, see Table 1. Schee p-values: *, signi®cance at >10%; ±, signi®cance at 10%; a, signi®cance at 5%; aa, signi®cance at 1% Variable
Spawning grounds (jigged)
Partially spent
Oshore
ML WT Ovary Oviduc FecOv FecOv4 FecOd GSI
aa * * aa a a aa;± a
aa a * aa * * ± aa
aa a * aa a a aa a; aa
317
duct. Squid collected oshore had signi®cantly higher (p < 0.01) GSI values than those inshore (Table 2). This is because squid jigged inshore had signi®cantly lower numbers of eggs in the oviducts (p < 0.01) and hence a lower oviduct weight (Table 1). Seasonal data for the inshore spawning grounds are set out in Tables 3 and 4. An increase in advanced oocyte numbers was found from autumn to spring, with low summer values, indicating an increase in gonad activity towards spring, the beginning of the peak spawning season. The winter sample also had the heaviest oviducts (Table 4). Table 5 gives the results of the aquarium experiments. Three individuals spawned twice within 12 to 36 h. One individual deposited an estimated 7916 eggs and had 5574 advanced oocytes left in the ovary after death.
Table 4 Loligo vulgaris reynaudii. Seasonal data: results of ScheeÂ's multiple comparison procedure. Schee p-values: ±, signi®cance at 10%; a, signi®cance at 5%; aa, signi®cance at 1% (Su summer; Au autumn; W winter; S spring; other abbreviations, see Table 1) Variable
ScheeÂ-test results
ML
Su; W = a Su; S = aa
WT
Su; Au = a Su; W = aa Su; S = aa
Ovary
Su; S = a W; S = ±
Oviduc
Su; Au = aa Su; W = aa S; Au = a S; W = aa
FecOv4
Su; S = a Au; S = a W; S = ±
FecOd
Su; Au = aa Su; W = aa S; Au = a S; W = aa
GSI
W; S = aa
Discussion McGowan (1954), Fields (1965) and Recksiek and Frey (1978) observed that Loligo opalescens concentrate in Table 3 Loligo vulgaris reynaudii. Seasonal data: fecundity and GSI values of squid collected on the spawning grounds. Abbreviations, see Table 1 Variable
Mean
SD
Min.
Max.
Summer N = 17 ML 185.12 WT 156.76 Ovary 6.89 Oviduc 1.58 FecOv 15585 FecOv4 7318 FecOd 617 GSI 5.68
25.07 62.82 2.18 2.36 4110 1879 1028 1.86
144.00 75.00 3.33 0.09 8894 3754 6 3.17
223.00 265.00 10.58 9.22 24118 10805 4056 9.88
Autumn N = 19 ML 199.47 WT 198.84 Ovary 7.61 Oviduc 4.43 FecOv 14664 FecOv4 7197 FecOd 1611 GSI 6.12
17.98 47.61 2.11 3.66 2621 1607 1382 1.82
165.00 122.00 4.78 0.26 10437 4605 48 3.17
229.00 290.00 12.10 13.16 18495 9722 4695 9.99
Winter N = 28 ML 201.32 WT 201.57 Ovary 7.63 Oviduc 8.16 FecOv 15185 FecOv4 8150 FecOd 2971 GSI 7.59
18.94 52.14 2.38 7.74 3980 2491 3017 3.05
146.00 92.00 1.63 0.11 6335 2185 19 1.88
232.00 308.00 13.01 29.11 25913 14094 11507 13.09
Spring N = 28 ML 212.82 WT 223.79 Ovary 9.43 Oviduc 1.86 FecOv 16459 FecOv4 9547 FecOd 608 GSI 5.03
13.48 33.94 2.18 2.24 3167 1813 837 1.26
185.00 145.00 5.02 0.23 9786 5410 37 2.92
233.00 285.00 12.55 7.26 23177 12733 3372 7.67
Table 5 Loligo vulgaris reynaudii. Aquarium maintenance: number of eggs deposited and the remaining number of oocytes in the ovary and oviduct [FecOv number of discernible oocytes in the ovary; FecOv4 number of oocytes greater than Stage 4 (Cowden 1968) in the ovary; FecOd number of mature eggs in the oviduct; n.d. no data] Female No. of egg strands No. of eggs FecOv FecOv4 FecOd number deposited deposited 1 2 3 4 5 5 6 7 8 8 9 9
5 30 25 47 47 8 18 18 42 17 35 15
740 4440 3700 6956 6956 1184 2664 2664 6212 1700 5180 1500
n.d. n.d. n.d. n.d. n.d. n.d. n.d. 13 868 n.d. 16 337 n.d. 15 278
n.d. n.d. n.d. n.d. n.d. n.d. n.d. 5 947 n.d. 5 574 n.d. 6 391
n.d. n.d. n.d. n.d. n.d. n.d. n.d. 483 n.d. n.d. 21
areas to spawn, followed by a mass mortality. Those studies were in¯uential in generating the idea of semelparity and synchronous spawning in squid populations. Maturation within a single breeding season has been suggested for several other loliginid species, e.g. Loligo pealei (Summer 1971, 1983; Macy 1980) and Loligo forbesi (Holme 1974; Martins 1982). However more recent studies on loliginids suggest that, although some species may exhibit fast growth, spawning is over a protracted period (Mangold 1987; Mangold et al. 1993). Other species of cephalopods, especially deep-sea and cold-water species, may grow more slowly, have a
318
prolonged spawning season and a longer life span (Mangold et al. 1993). For example, Jackson (1993) found that slower-growing, cool-season (spring) Idiosepius pygmaeus lived longer and had comparatively larger gonads than their warm-season counterparts, despite no dierence in body size between the two seasons. The methods used to calculate fecundity in this study are similar to those used by Vovk (1972) for Loligo paelei and by Durward et al. (1979) for Illex illecebrosus. A count of mature eggs in the ovary and oviduct of prespawning squid that spawn once in their lifetime should prove adequate in determining fecundity (Voss 1983). Various authors have pointed out that, in species likely to spawn over a protracted period, this method may well yield an underestimate (e.g. Mangold 1987; Coelho et al. 1994; Collins et al. 1995). Melo and Sauer (1999) con®rm that Loligo vulgaris reynaudii ®ts into the latter category, complicating the estimation of egg production. Apart from the results of the aquarium experiments, serial spawning is also suggested from this study where the partially spent group (obtained after the deposition of an egg strand) had signi®cantly larger numbers of mature eggs in the oviduct than squid jigged above the egg mass. As mentioned previously, it is possible that squid go through a resting phase (minutes or hours) between successive spawning bouts of one spawning site, or before moving to other spawning areas. During this time they may be more susceptible to the jigs. Tagging studies have shown spawning squid to move between spawning sites over a period of weeks (Sauer et al. 1999). Estimations of egg production in loliginids vary widely. For example, Mangold-Wirz (1963) showed that the fecundity of Loligo vulgaris does not exceed 6000 eggs and in most cases varied between 3000 and 5000 eggs. Vovk (1972) counted the number of eggs >0.5 mm in the ovary and found a range of between 2500 and 15 900 for Loligo pealei, whereas Collins et al. (1995) estimated potential fecundity for Loligo forbesi to be between 2500 and 10 500. One estimation of actual fecundity for L. v. reynaudii can be taken from the aquarium experiments, where 8140 eggs were spawned over 36 h. However the estimate of potential fecundity given in the results is in the region of 17 000 eggs. A true ®gure of actual fecundity therefore lies somewhere between the two values. More precise methods for calculating actual fecundity are required. Although serial spawning has been proposed for a number of species, the number of batches of eggs spawned from a particular ovary, the number of eggs released per batch and frequency of spawning have not been determined. Analysis of the rate of resorption of post-ovulatory follicles in the ovaries of squid collected at regular intervals over one spawning cycle should provide this information. Melo and Sauer (1998) have shown that atretic oocytes are present in both developing and mature ovaries in Loligo vulgaris reynaudii. In fact, spent individuals can be identi®ed from the high numbers of atretic oocytes
present. Some oocytes are therefore destined not to mature, further complicating fecundity estimates. Laptikhovsky and Nigmatullin (1993) found similar results for Illex sp., and Nigmatullin and Laptikhovsky (1994) calculated actual fecundity to be only 60 to 80% of the potential fecundity in the Ommastrephidae. To the authors' knowledge, no estimates of this eect on fecundity are available for Loliginidae. Follicular atresia will also lower the number of maturing eggs in the ovary. To correct for this it is necessary to estimate the duration of each atresia stage in relation to the total spawning period (Witthames and Greer-Walker 1991). Samples collected over a complete spawning season may answer this question, allowing for a more precise calculation of actual fecundity in loliginids. In the present study potential fecundity in Loligo vulgaris reynaudii was found to be higher on the inshore spawning grounds than oshore. This may be attributable to resorption. The lower number of eggs in the oviducts of squid inshore implies more frequent spawning on the inshore spawning grounds. As mature females on the oshore feeding grounds have oviducts ®lled with mature eggs it is therefore likely that L. v. reynaudii store mature eggs for extended periods of time without spawning. Ikeda et al. (1991) made similar observations for Todarodes paci®cus. This implies that L. v. reynaudii are not generally forced to empty their oviducts of eggs in deeper water during periods of unfavourable environmental conditions inshore, a theory proposed to explain the eggs of L. v. reynaudii caught in trawl nets in depths of water greater than 60 m. In conclusion, future investigators will have to make use of innovative techniques to solve some of the complex problems involved in fecundity estimation in loliginid squid. Where possible, research programmes should aim to collect regular samples over a complete spawning season. Acknowledgements We wish to thank Mr D. Venter of the University of Port Elizabeth for his assistance with the statistical analysis, Dr M. R. Lipinski for his valuable comments on various drafts of this paper, and the master, crew and technicians on the F. R. S. ``Africana'' for assistance in collecting the gonad samples. Sea Fisheries, Cape Town and the South African Squid Management Industrial Association (SASMIA) are gratefully acknowledged for the funding of the project. The Small Business Development Corporation sponsored a skiboat and Caltex Oil, S.A., a Land Rover and fuel for the boat and vehicle. The Port Elizabeth Museum housed the project.
References Augustyn CJ, Lipinski MR, Sauer WHH, Roberts MJ, MitchellInnes BA (1994) Chokka squid on the Agulhas Bank: life, history and ecology. S Afr J mar Sci 90: 143±154 Badenhorst JH (1974) The morphology and histology of the male genital system of the squid Loligo reynaudii (d¢Orbigny). Ann Univ Stellenbosch (Ser A) 49(1): 1±36
319 Boyle PR, Pierce GJ, Hastie LC (1995) Flexible reproductive strategies in the squid Loligo forbesi. Mar Biol 121: 501±508 Coelho ML, Quintela J, Bettencourt V, Olavo G, Villa H (1994) Population structure, maturation patterns and fecundity of the squid Loligo vulgaris from southern Portugal. Fish Res 21: 87± 102 Collins MA, Burnell GM, Rodhouse PG (1995) Reproductive strategies of male and female Loligo forbesi (Cephalopoda: Loliginidae). J mar biol Ass UK 75: 621±634 Cowden RR (1968) Cytological and cytochemical studies of oocyte development of the follicular epithelium in the squid, Loligo brevis. Acta Embryol Morph exp 10: 160±173 Durward RD, Amaratunga T, O'Dor RK (1979) Maturation index and fecundity for female squid, Illex illecebrosus (LeSueur, 1821). Res Bull int Commn NW Atlant Fish 14: 67±72 Fields WG (1965) The structure, development, food relations, reproduction and life history of the squid Loligo opalescens Berry. Fish Bull Calif 131: 1±108 Harman RF, Young RE, Reid SB, Mangold KM, Suzuki T, Hixon RF (1989) Evidence for multiple spawning in the tropical oceanic squid Stenoteuthis ovalaniensis (Teuthoidea: Ommastrephidae). Mar Biol 101: 513±519 Hat®eld EMC, Rodhouse PG, Porebski J (1990) Demography and distribution of the Patagonian squid (Loligo gahi d¢Orbigny) during the austral winter. J Cons int Explor Mer 46: 306±312 Hixon RF (1983) Loligo opalescens. In: Boyle PR (ed) Cephalopod life cycle. 1. Species accounts. Academic Press, London, pp 95± 114 Holme NA (1974) The biology of Loligo forbesi Steenstrup (Mollusca: Cephalopoda) in the Plymouth area. J mar biol Ass UK 54: 481±503 Ikeda AY, Sakurai Y, Shimazaki K (1991) Development of female reproductive organs during sexual maturation in the Japanese common squid Todarodes paci®cus. Nippon Suisan Gakk 57(12): 2243±2247 Jackson GD (1993) Seasonal variation in reproductive investment in the tropical loliginid squid Loligo chinensis and the small tropical sepioid Idiosepius pygmaeus. Fish Bull US 91: 260±270 Laptikhovsky VV, Nigmatullin CM (1993) Egg size, fecundity, and spawning in females of the genus Illex (Cephalopoda: Ommastrephidae ). ICES J mar Sci 50: 393±403 Lewis AR, Choat JH (1993) Spawning mode and reproductive output of the tropical cephalopod Idiosepius pygmaeus. Can J Fish aquat Sciences 50: 20±28 Lipinski MR (1979) Universal maturity scale for the commerciallyimportant squids (Cephalopoda: Teuthoidea). The results of maturity classi®cation of the Illex illecebrosus (LeSueur, 1821) populations for the years 1973±1977. Res. Doc. int. Commn NW. Atl Fish 79/II/38, Dartmouth, Canada Lipinski MR, Underhill G (1995) Sexual maturation in squid: quantum or continuum? S Afr J mar Sci 15: 207±223 Lum-Kong A, Pierce GJ, Yau C (1992) Timing of spawning and recruitment in Loligo forbesi (Cephalopoda: Loliginidae) in Scottish waters. J mar biol Ass UK 72(2): 301±311 Macy WK (1980) Development and application of an objective method for classifying long-®nned squid, Loligo pealei, into sexual maturity stages. Fish Bull (Wash DC) 80(3): 449±459
Mangold K (1987) Reproduction. In: Boyle PR (ed) Cephalopod life cycles. 2. Comparative reviews. Academic Press, London, pp 157±200 Mangold KM, Young RE, Nixon M (1993), Growth versus maturation in cephalopods. In: Okutani T, O'Dor RK, Kubodera T (eds) Recent advances in cephalopod ®sheries biology. Tokai University Press, Tokyo, pp 697±703 Mangold-Wirz K (1963) Biologie des cephalapodes benthiques et nectoniques de la mer Catalagne. Vie Milieu 13(Suppl): p256 Martins HR (1982) Biological studies of the exploited stock of Loligo forbesi (Mollusca: Cephalopoda) in the Azores. J mar biol Ass UK 62: 799±808 McGowan J (1954) Observations on the sexual behaviour and spawning of the squid, Loligo opalescens at La Jolla, California. Calif Fish Game 40(1): 47±54 Melo YC, Sauer WHH (1998) Ovarian atresia in cephalopods. S Afr J mar Sci 20: 143±151 Melo YC, Sauer WHH (1999) Con®rmation of serial spawning in the chokka squid Loligo vulgaris reynaudii o the coast of South Africa. Mar Biol 135: 307±313 Moltschaniwskyj NA (1995) Multiple spawning in the tropical squid Photololigo sp.: what is the cost in somatic growth? Mar Biol 124: 127±135 Nigmatullin ChM, Laptikhovsky VV (1994) Reproductive strategies in the squids of the family Ommastrephidae (preliminary report). Ruthenica 4(1): 79±82 Recksiek CW, Frey HW (1978) Biological oceanographic and acoustic aspects of the market squid, Loligo opalescens Berry. Fish Bull Calif 169: 1±185 Sauer WHH, Lipinski MR (1990) Histological validation of morphological stages of sexual maturity in chokker squid Loligo vulgaris reynaudii D¢Orb (Cephalopoda: Loliginidae). S Afr J mar Sci 9: 189±200 Sauer WHH, Lipinski MR, Augustyn CJ (1999) Tag recapture studies of the chokka squid Loligo vulgaris reynaudii d'Orbigny, 1845 on inshore spawning grounds on the south east coast of South Africa. Fish Res (in press) Sauer WHH, McCarthy C, Smale MJ, Koorts AS (1993) An investigation of the egg distribution of the chokka squid, Loligo vulgaris reynaudii, in Krom Bay, South Africa. Bull mar Sci 53(3): 1066±1077 Summers WC (1971) Age and growth of Loligo paelei, a population study of the common Atlantic coast squid. Biol Bull mar biol Lab, Woods Hole 141: 189±201 Summers WC (1983) Loligo pealei. In: Boyle PR (ed) Cephalopod life cycles. 1. Species accounts. Academic Press, London, pp 115±142 Voss AN (1983) A review of cephalopod ®sheries biology. Mem natn Mus Vict 44: 229±241 Vovk AN (1972) Fecundity of the North American squid Loligo pealei (LeSueur, 1821). Trudy Atlant nauchno-issled Inst morsk ryb khoz Okeanogr 42: 133±140 (in Russian) (FRB Transl Ser No 3302) Witthames PR, Greer-Walker M (1991) Follicular atresia in the ovary of the western mackerel (Scomber scombrus L.). In: Scott AP, Sumpter JP, Kine DE, Rolfe MS (eds) Proceedings of the 4th International Symposium on the Reproductive Physiology of Fish, University of East Anglia, Norwich, July 1991. Fish Symposium, Sheeld, pp 1±328