Marine Biology $2002) 140: 165±173 DOI 10.1007/s002270100678 A.T. Correia á C. Antunes á J. Coimbra Aspects of the early life history of the European conger eel Conger conger ) inferred from the otolith microstructure of metamorphic larvae Received: 2 March 2001 / Accepted: 13 July 2001 / Published online: 2 October 2001 Ó Springer-Verlag 2001 Abstract The life cycle of the conger eel $Conger conger shores of the Atlantic coast of Europe from Norway to L.) is still fairly unknown, especially its larval lepto- the Mediterranean, and extending into the western Black cephalus phase. The morphology, biometry and meristic Sea $Bauchot and Saldanha 1986). In spite of its rela- characteristics of 184 metamorphosing conger eel larvae tively wide distribution, its life cycle is poorly known, collected from the Minho River, northern Portugal, be- especially during its leptocephalus phase. tween December 1998 and April 1999 were analysed. The Schmidt $1931) caught small conger eel larvae in the total number of myomeres, the general body morphology Sargasso Sea, Mediterranean and Northeast Atlantic, and the pigmentation pattern of leptocephali are in suggesting a similar migratory behaviour to the Euro- agreement with the corresponding data found by other pean eel $Anguilla anguilla L.), i.e. spawning in the authors for this species. The sagitta microstructure of 90 Sargasso Sea, following a larval transoceanic migration specimens was viewed by scanning electron microscopy. to the European coast. However, the larvae caught in The numbers of daily increments obtained in the otolith's the Sargasso Sea and wrongly identi®ed as C. conger, countable zone were between 205 and 324. The ®nal part were indeed C. triporiceps $McCleave and Miller 1994). of the countable zone is characterised by a sharp increase Nowadays the larvae of both species are distinguished in width of the daily increments $transition zone), fol- by the catch location, being separated by a line that goes lowed by a peripheral diuse zone. In the diuse zone no through the Canary Islands and the western zone of the narrow circumscribed rings are visible, which prevents an Azores Islands, in a NW direction. The European con- accurate estimate of the duration of metamorphosis. The ger eel leptocephali are restricted to the central and data indicates that the size variation of metamorphosing eastern zones of the Atlantic Ocean $Strehlow et al. leptocephali is large, suggesting that their hatching time 1998). must be variable. Our data also show that the largest Dierent spawning places have been suggested for larvae, in later stages of metamorphosis, arrive ®rst to the C. conger. Lythgoe and Lythgoe $1971), Bagenal and northern Portuguese coastal waters. Kenney $1973) and Wheeler $1985) found that conger eel spawn only once, at great depths $3000±4000 m), during the summer, in the Northeast Atlantic between Gibral- Introduction tar and the Azores. There are also spawning areas in the Mediterranean $Wheeler 1985). However, until now, The European conger eel $Conger conger L.) is a marine the only spawning area well known for this species is in benthic ®sh, commonly found along sandy and rocky the central-east basin of the Mediterranean $Cau and Manconi 1983). It has been suggested that the leptocephali has a long Communicated by S.A. Poulet, Rosco larval life $Bauchot and Saldanha 1986), taking about 1 or 2 years to drift inshore and to reach the juvenile elver A.T. Correia $&) á J. Coimbra form $Lythgoe and Lythgoe 1971; Wheeler 1985; Instituto Cieà ncias Biome dicas Abel Salazar, Largo Abel Salazar 2, 4099-033 Porto, Portugal Strehlow 1992). Recent studies, based on morphometric and sagitta analysis of premetamorphic larvae, showed E-mail: [email protected] Tel.: +351-2-26080477 that spawning occurs in the Mediterranean Sea between Fax: +351-2-26060423 July and September. After a short growth period, the larvae $>30 mm) start migration around November, in A.T. Correia á C. Antunes á J. Coimbra Centro Interdisciplinar de InvestigacË aÄ o Marinha e Ambiental, a NW direction toward southern Portugal and Spain, Rua do Campo Alegre 823, 4150-180 Porto, Portugal extending throughout the eastern and central zones of 166 the Atlantic. The conger eel has a second growth period, lasting until the beginning of the next summer $normally reaching 130±150 mm, with a maximum of 165 mm length), after which they start migration in the direction of the coastal waters of the continental slope, with a possible return to the Mediterranean. It is supposed that this coastal migration induces metamorphosis $Strehlow et al. 1998). However, the location and timing of meta- morphosis is unknown. The microstructure of the otoliths provides us with valuable information about the early life cycle of ®shes. Strehlow et al. $1998) obtained a mean value of 280 days by analysis of the sagittae of premetamorphic lepto- cephali $between 80 and 120 mm long), using the clear daily increments from the nucleus to the otolith margin $Antunes 1994; Antunes and Tesch 1997). However, Fig. 1 Conger conger. Sampling location of the metamorphosing conger eel leptocephali until now the sagitta microstructure of metamorphic larvae has not been described. The present paper examines the relationships between the biometry/meristic data and the sagitta microstruc- tural growth in C. conger leptocephali during meta- morphosis, in an attempt to elucidate some aspects of its larval life history. Materials and methods The conger eel larvae $Conger conger L.) were obtained as by-catch from the glass eel ®shery at the mouth of the Minho River in northern Portugal $Fig. 1), monthly between December 1998 and April 1999. Fishing, using a stow net, took place in the estuarine area during the night, in the period of the new moon during the ¯ood-tide current $Antunes 1994). After collection, the leptocephali $n=184) were preserved in commercial ethanol $95%) and the general body morphology, pigmentation, morphometric and meristic characters $number of myomeres) were analysed according to the methodology described by Smith $1989). Because the preservation method induces shrinkage of the body $approximately 7.5%), the length measure- Fig. 2 Conger conger. Schematic diagram of the otolithometric ments were corrected. measurements in the sagitta $D maximum diameter; R maximum Both sagittae were removed from 90 specimens $randomly se- radius; CZW maximum width of the countable zone; DZW lected), cleaned, mounted on cylindrical stubs and polished with maximum width of the diuse zone; C core; FFC ®rst feeding 2400 silicon carbide abrasive paper and aluminium paste until the check) core was revealed. After that, they were etched for 8 s with a 0.5% solution of HCl, sputter-coated with gold under vacuum and viewed with scanning electron microscope $SEM; Jeol JSM 630-1F) The percentage of the area occupied by the accessory growth at 15 kV. centres $%AGCA) of the total otolith surface was also calculated. Following SEM analysis, core diameter $CD), maximum otolith Statistica 5.0 was used to perform all statistical analyses. After diameter $D), maximum otolith radius $R), maximum width of the testing for normality and homogeneity of variances, a one-way countable incremental zone $CZW) and maximum width of the analysis of variance $ANOVA) was used to examine the dierences diuse zone $DZW) were measured from the SEM photographs in the relationships between morphometric and otolithometric $Fig. 2). measures among sampling periods. ANOVA was followed by a We considered the growth increment in the larval otolith of Tukey HSD-test for unequal n $Spjotvoll/Stoline test). Data are conger to be daily, although daily deposition has not been validated presented as meansstandard deviations $SD). in this species. We base this assumption on the results of several related species, e.g. Conger myriaster $Mochioka et al. 1989), Anguilla japonica $Umezawa et al. 1989; Umezawa and Tsukamoto 1991) and A. rostrata $Martin 1995), which have been shown to Results have daily depositions. In the countable zone $CZ), the total number and width of daily Morphometric and meristic data increments were registered. The average of every ten increment widths from the ®rst feeding check to the edge of the CZ was used to measure the otolith growth rate. The maximum axis of the CZ The conger eel $Conger conger) larvae were identi®ed was regarded as the otolith radius, along which increment widths mainly by counting myomeres. Unfortunately, because were measured. some specimens were damaged from the preservation 167 Table 1 Conger conger. Morphometric and meristic counts of The metamorphic larvae had the body shape associ- metamorphic conger eel larvae $SD standard deviation; n sample ated with the general and identifying features of this size; TNM total number of myomeres; PDM predorsal myomeres; PAM preanal myomeres; PDL predorsal length; TL total length; species $Fig. 5). They had a laterally compressed, PAL preanal length; BD body depth; HL head length) transparent body, with W-shaped myomeres and a simple tubular gut along the ventral margin of the body. Parameter Range MeanSD n The head was of medium size and had rounded eyes. TNM 154±163 89 Long dorsal and anal ®ns were con¯uent with a rounded PDM 25±51 82 caudal ®n. Larvae had small pectoral ®ns, and pelvic ®ns PDM/TNM ± 0.230.04 were absent. PAM 47±65 81 Concerning the pigmentation, they possessed large PAM/TNM ± 0.360.03 PDL $mm) ± 37.355.97 146 dots along the lateral line, which became sparser or PDL/TL ± 0.29+0.04 disappeared anteriorly. We also observed some mela- PAL $mm) ± 55.335.97 153 nophores at the bases of the caudal and anal ®n rays, PAL/TL ± 0.430.04 and limited to posterior region of the dorsal ®n. They BD $mm) ± 7.481.35 142 HL $mm) ± 9.500.55 184 also exhibited a double ventral pigmentation on the sides of the intestine, sometimes extending slightly beyond the anus.