Mediterranean Marine Science Vol. 3/2, 2002, 147-158 Age determination and growth of leaping mullet, (Liza saliens R.1810) from the Messolonghi Etoliko lagoon (western Greece). G. KATSELIS1,2, C. KOUTSIKOPOULOS1 and P. KASPIRIS1 1Department of Biology, University of Patras, 26500, Patra Greece 2 Department of Aquaculture and Fisheries Technology, Academic Technological Educational Institute Messolonghiou, 30200 Messolonghi, Greece e-mail: gkatselis@yahoo.com Abstract This study is the first detailed work on the age and growth of the leaping mullet (Liza saliens, Risso 1810) in the central Mediterranean. During the period 1991-1995 the age and growth of leaping mullet from the Messolonghi -Etoliko lagoon system (western Greek coast) were studied. Age and growth determinations were based upon otolith samples taken from 537 fish. Marginal increment analysis was used to validate age determination. Annulus formation took place around November each year. The back-calculated lengths at age estimated from the otoliths showed no differences between sub-areas of the lagoon system and the recorded limited between-years variability showed no persistent temporal pattern. The maximum age of leaping mullet in the Messolonghi - Etoliko lagoon was 5 years for males and 6 years for females. The von Bertalanffy equation (L∞=32.99±1.25 cm, k=0.258 ±0.017 year-1, t0=-0.47±0.04 year) accurately describes the growth of the total length of leaping grey mullet for all life stages (fry, juveniles and adults). A large spread and length overlap characterized the age groups. The estimated Length-Weight relationships were common for the two sexes (W=0.0079L3.01). Keywords: Liza saliens, Mugilidae, leaping mullet, age and growth, Western Greece. Introduction The leaping mullet (Liza saliens, Risso 1810, Pisces: Mugilidae) is abundant around the Mediterranean, the Black Sea coast and along the Atlantic coast from Morocco to the Bay of Biscay (TREWAVAS, 1979). It is also present in the Caspian Sea (BEN TUVIA, 1986) and in the brackish water lakes of Egypt (EL ZARKA & KAMEL, 1965). Medit. Mar. Sci., 3/2, 2002, 147-158 Mullet and mullet products have considerable economic importance at a regional level, especially around the Mediterranean. The leaping mullet constitutes an important part of their production. It is one of the least studied Mugilids in the central and eastern Mediterranean. Information on its biology comes mainly from studies carried out in the western Mediterranean (CAMBRONY, 1983), southeastern Mediterranean (ZAKY- 147 RAFAIL, 1968; El ZARKA & EL SEDFY, 1970; EL ZARKA et al., 1970; PELMUTTER et al., 1957) and eastern Mediterranean (KOUTRAKIS & SINIS, 1994; KOUTRAKIS et al., 1994; KAYA et al., 2000). The other references to this species are from the Black Sea (ALEEV, 1956-cited by THOMSON, 1966), the european Atlantic coast (DRAKE et al., 1984 a,b) and the Balearic Islands (CARDONA, 1999 a,b). Most of this information concerns age, growth, morphological character, feeding and seasonal occurrence. The studies conducted in the central Mediterranean examine morphological characters (MINOS et al., 1994; MINOS et al., 1995; MINOS et al., 2002), seasonal occurrence of fry, the spawning period of the species (KATAVIC, 1980; KATSELIS et al., 1994; MICKOVIC et al., 1994), feeding habits (TORRICELLI et al., 1988; TOSI and TORRICELLI, 1988) and dynamic of population (CONIDES et al., 1992). In this region, only one study presented information on the age and growth (GIATNISI, 1985), but the estimates were of low accuracy. Age and growth are important input parameters for stock assessment techniques. This work is the first detailed study on these parameters of leaping mullet in the Messolonghi-Etoliko lagoon, a typical Mediterranean lagoon, (western Greece, central Mediterranean). Materials and Methods The Messolonghi Etoliko lagoon system is located along the coast of western Greece (38Æ18’N-21Æ32’E), and is situated in the northern part of the Patraikos Gulf. It has a total surface area of about 150 KmÇ and is one of the larger lagoon systems in the Mediterranean. Based on the topography, hydrology and fish species composition, six sub-areas can be defined (DIMITRIOU et al., 2000a) (Fig.1). Except for its northern part (Etoliko lagoon), the lagoon has an average depth of 0.8m and the bottom is covered with Fig 1: The area of study. Letters R1 to R6 indicate the different areas composing the MessolonghiEtoliko lagoon system. 148 Medit. Mar. Sci., 3/2, 2002, 147-158 rich vegetation. The Etoliko lagoon has an average depth of 12m and a maximum depth of 33m. The water temperature ranges from 9-32Æ C (DIMITRIOU et al., 2000b) and the salinity from 15-45 psu. Exploitation is a common, extensive culture, based on seasonal entrance to the lagoons and summer-to-winter offshore fish migration. The fish are caught in barrier traps in fenced areas separating the lagoon from the sea. Gill nets and dip nets are also used in the lagoon all year long. During the period from January 1991 to December 1995, samples of leaping grey mullet were collected on a monthly basis from four sub-areas (R2, R3, R5 & R6) of the Messolonghi-Etoliko lagoon system (Fig. 1). A total sample of 1133 specimens was caught using a variety of different types of equipment. 973 specimens were caught in barrier traps, 137 specimens by trammel nets and 23 specimens with total length of less than 100 mm, by 12 x 1.2 m mosquito purse seine. Total length (L) and standard length (LS) were measured to the nearest mm, total body weight (W) to the nearest 0.01 g and the sex was determined in mature specimens by gonad colour: orange for females and white for males, while otoliths were removed from the fish. Age determination from otolith readings was performed on 537 randomly collected fish monthly. Otoliths were read under a binocular stereoscope with a 25x magnification, using transmitted light. The opaque zone was used for ageing, the distances from the otolith nucleus to the rostrum (maximum radius), and the radius (distance from the otolith nucleus to the opaque zone edge, in the same direction of maximum radius) of each ring were measured with an optical micrometer. In otoliths presenting reading difficulties in the region near the focus (mainly from adult individuals), matter was removed from the surface layers of the bezel side of the otolith by smoothing with 50Ì sandpaper. The marginal increment index I M = 100*(Rc-Rn)/Rn, was calculated, where Rn is Medit. Mar. Sci., 3/2, 2002, 147-158 the radius of the last growth ring n and Rc is the maximum otolith radius. Relationships between total otolith radius (Rc) and total length (L) were determined for males, females and the sexes combined. Regression analysis was employed and the significance of the recorded differences was tested by an analysis of covariance (ANCOVA). Back-calculated lengths at the estimated age of individual fish were estimated from the log transformation length-otolith radius regression. Differences in backcalculated length at age groups between the age, the sexes, years of sampling and sub-areas were tested by analysis of variance (ANOVA) (ZAR, 1984). The von Bertalanffy growth equation was fitted to back-calculated length at age data using a non Linear Regression Analysis. Results The 1133 leaping mullet collected during the period 1991-1995 ranged from 4 to 29.1 cm total length. From these, 268 individuals were identified as male, 216 as female and in 649 immature individuals sexes were not identified. The overall ratio of males to females was 1: 0.8 and ¯2 analysis (P>0.05) revealed this to be not significantly different from the expected 1:1 ratio. Females ranged from 15 to 29.1 cm total length and males from 12 to 24.1 cm (Fig. 2). The regression equations of the interconversion of L to LS are expressed linearly and the slopes of the equation did not differ significantly between the sexes (F=0.16, P>0.05). The relationship for the sexes combined is: L=0.456+1.234LS (RÇ=0.96, SEest=0.74) where SEest is the standard error of estimates. The analysis of covariance showed no significant difference between the males and females in the Length-Weight relationship (F=0.14, P>0.05). The obtained equation for both sexes is the following: W=0.0079L3.01 149 (RÇ=0.97, SEest=0.11), where W is expressed in g and L in cm. Age and growth determinations were based upon otolith samples of 537 fish. Ring reading was impossible in 42 (7.9%) of them. The unreadable otoliths were uniformly distributed throughout the fish size range. The maximum number of detected rings was 6 for males and 7 for females. The mean marginal increment index for the age group II, III and IV showed that one ring was formed per year. The largest percentage of individuals with a ring at the edge of the otolith was found in the November samples (Fig. 3). The analysis of covariance showed no significant difference between the males and females in the L-Rc relationships (F=4x10-6, P>0.05). Thus, for the estimation of backcalculated lengths at each age (Table 1) we used the relationships obtained from the total sample (ln(L)=-2.13+1.19 ln(Rc), RÇ=0.98, SEest=0.083). Analysis of variance confirms that for the age groups O to V, there are no significant differences between age groups and sexes for the back-calculated mean length at each age from otoliths (P>0.05). The first ring was formed in otoliths at an average length of 3.8 ±1.32 cm. The spawning period of leaping Fig 2: Total length (L) frequency distribution by sex for the sampled leaping mullet individuals. Fig 3: Monthly changes in the marginal increment index (IM) estimated from the age groups II, III and IV. 150 Medit. Mar. Sci., 3/2, 2002, 147-158 Table 1 Average back calculated lengths (cm) for each ring in each age group. Males Females Age group O I II III IV V VI F P Mean SD n growth Mean SD n growth Mean SD n growth F P 1 3.53 3.74 3.71 3.86 4.91 4.72 2.1 0.06 3.81 1.32 198 11.44 3.7 1.5 39 11.19 4.08 1.52 7 12.23 1.37 0.24 2 3 Otolith rings 4 5 10.81 10.38 10.42 10.55 10.19 10.04 1.51 0.18 10.44 1.09 454 6.63 10.4 1.1 82 6.7 10.44 1.06 145 6.36 0.041 0.84 15.68 15.56 15.77 15.74 15.50 0.34 0.84 15.65 1.35 415 5.21 15.8 1.4 81 5.4 15.84 1.09 146 5.40 2.3 0.63 19.57 19.37 19.69 19.48 0.49 0.68 19.53 1.20 225 3.88 19.5 1.5 26 3.7 19.54 1.03 113 3.70 0.09 0.75 22.37 22.83 22.71 0.88 0.42 22.50 1.40 82 2.97 21.0 1.5 4 1.5 22.46 1.47 41 2.92 3.85 0.06 6 25.49 24.57 3.43 0.07 25.38 0.84 24 2.88 25.1 1 4.2 25.38 1.03 13 2.92 0.059 0.81 7 26.58 26.58 0.97 3 1.20 25.46 1 0.07 n is the number of otolith readings, SD is standard deviation, F is F-ratio and P is significant level. mullet begins in May-June and the greatest frequency of specimens with a total length between 35-70 mm appeared in the coastal waters of Messolonghi from October December (KATSELIS et al., 1994). Thus, it can be assumed that the first ring is formed 45 months after spawning and consists of the ‘fry ring’. The individuals that have only this ring formed the O age group. From the results presented above and the marginal increment analysis results, the 1st ring is formed on the otoliths at an age of 4-5 months, the 2nd at 1617 months and the rest follow at 12 monthly regular intervals. The back-calculated lengths at age between years also presented some variability. In fact, an analysis of variance showed that for the age groups I to III there were significant differences between years (Table 2), but these differences do not show a persistent temporal pattern. In Medit. Mar. Sci., 3/2, 2002, 147-158 fact, the smaller group I individuals sampled in 1994 produce the longer mean values the next year. Also, ANOVA confirms that for age groups O to V, there are no significant differences between sub-areas (P>0.05) for the back-calculated mean length at each age from otoliths. The growth rate was calculated using the von Bertalanffy growth equation and the coefficients of the growth are presented in Table 2. The 95 % confidence intervals of the von Bertalanffy growth equation parameters overlap between the years of sampling, which means that their parameters showed no significant differences. The ratio of Lmax/L∞ (Lmax is the maximum observed length in the sample) was 0.88. The estimated value of t0 was –0.47±0.032 year (5.18-6.1 months) and can be considered as the time needed to form the first ring (‘fry ring’) in otoliths. 151 Table 2 Size and age composition of males, females and immature individuals of leaping mullet from Messolonghi -Etoliko lagoon. Age group O I II III IV V N Observed range of L L∞ (cm) k (year-1) t0 (year) n RÇ 1991 3.09 10.4 15.28 19.07 22.45 25.08 44 1992 4.26 10.7 15.68 19.61 22.9 25.42 113 11.1-29.1 4.5-28.4 31.8±5.02 35.1±3.7 0.27±0.09 0.23±0.05 -0.44±0.17 -0.57±0.12 122 278 0.94 0.93 year of sampling 1993 3.36 10.13 14.9 18.9 21.43 25.14 102 4.2-28.7 32.5±3.07 0.25±0.03 -0.45±0.08 228 0.95 1994 4.18 9.9 15.03 18.92 23.76 26.42 29 1995 3.88 10.49 15.97 19.75 22.41 25.69 205 n F P 197 1.82 0.13 450 5.048 0* 411 11.05 0* 221 5.068 0* 78 2.46 0.04* 20 0.54 0.71 12.8-23 47.2±23.9 0.14±0.1 -0.64±0.34 69 0.92 11.5-26.6 32.37±1.77 0.27±0.03 -0.46±0.04 705 0.95 All sampling 32.99±1.25 0.26±0.017 -0.47±0.03 1401 0.95 * Significant difference, N is number of specimens, n is the number of length-age readings for back calculation, F is F-ratio and P is significant level. A length-age key for the two sexes and their ramada (5.7%) and Chelon labrosous combination is presented in Table 3. The age (2.4%)(KATSELIS, unpublished data). groups were characterised by a large spread Various methods are used to determine the and overlapping of length. age in Mugilidae. Some authors preferred using length frequency analysis (DRAKE et al., 1984a; CARDONA, 1999a), but QUIGNARD and Discussion FARRUGIO (1981) claimed that these Studies of the age and growth of fish are methods were useless for mullet because of the essential for the management of their stocks. difficulty of obtaining samples, and because Grey mullet make up the most important part spawning takes place over several months. Both scales and otoliths are used for the of the fish catch in the Messolonghi - Etoliko age determination of Mugilidae. To use otolith, lagoons. The mean annual production is about or any other structure, for age determination, 1300 – 1500 mt and is provided by 200 the deposition of regular detectable age marks fishermen working at the barrier traps and 700 is essential. Otolith age determination is fishing in the lagoon. Mullet are estimated to compose 45 % of the total fish catch considered to be more accurate because (KOTSONIAS, 1984; DIMITRIOU et al., otoliths have a higher priority in utilization of 1994). Separate catch statistics for each species calcium (CARLANDER, 1987). However the are available only for barrier trap fish catches. otoliths have rarely been used in mullet because During the period 1995-1998, 182 mt of fish of the difficulty in reading the rings in the region per year was caught in the barrier trap of the near the focus of the otolith (ERMAN, 1959; Messolonghi -Etoliko lagoons, 51.9 % of which KENNEDY & FITZMAURICE, 1969; were mullet. Among the five mullet species BRULHET, 1974). THONG (1969) indicated caught barrier trap, L. saliens, Liza aurata and that only 67%, 66% and 64% of the otoliths of Mugil cephalus were more abundant (18.8%, L. ramada, C. labrosus and L. aurata respectively 14.3 % and 10.7 % respectively) than Liza were readable in the above-mentioned studies. 152 Medit. Mar. Sci., 3/2, 2002, 147-158 Table 3 Size and age composition of males, females and immature individuals of leaping mullet from Messolonghi -Etoliko lagoon. L (cm) I 4-5 5-6 6-7 7-8 8-9 9-10 10-11 11-12 12-13 13-14 14-15 15-16 16-17 1 17-18 1 18-19 19-20 20-21 21-22 22-23 23-24 24-25 25-26 26-27 27-28 28-29 29-30 Total 2 II males III IV V II females III IV V VI O 10 3 10 6 5 4 1 3 12 14 12 10 3 2 3 3 8 6 1 1 2 6 11 9 4 1 1 1 1 1 1 2 8 14 19 15 7 3 2 1 5 5 5 9 2 1 1 56 22 3 1 33 72 28 This difficulty was verified in the present study and it was partially overcome by smoothing the bezel side during the preparation of the otolith. The percentage of reliability of otolith readings in this study was 92.1%. Following the same technique KAYA et al. (2000) have increased the percentage of reliability of otolith readings to 87.7%. According to our findings, otolith readings gave correct results and fitted the growth equation well(Table 2). The results of the study of the marginal increment index suggest that rings in leaping mullet are formed annually around November, when the water temperature in the lagoon is between 14 and 21ÆC (DIMITRIOU et al., 2000b). Thus, the annual ring formation is related to the decrease of the growth rate of the species in water temperatures lower than Medit. Mar. Sci., 3/2, 2002, 147-158 2 1 3 2 3 1 12 I 2 2 5 1 5 5 9 7 3 immature II III IV 2 8 11 16 23 16 8 14 5 1 1 8 9 10 6 7 6 1 1 39 39 103 V 48 3 4 5 5 3 5 1 1 1 1 2 1 2 1 27 8 VI total 10 3 10 6 5 2 6 6 3 13 16 29 43 38 39 55 53 46 35 27 20 11 7 1 7 1 5 1 2 496 20ÆC (CARDONA, 1999a) and minimal feeding activity (CARDONA, 1999b). In addition, ‘fry ring’ formation is possibly caused by changes in the proteinic composition due to the different diet that they adopt (from zooplankton to mixed and finally to vegetarian) (ALBERTINI-BERHAUT, 1974; DE SILVA, 1980). Other researchers agree that the formation of the annual ring of mullet takes place in winter, but in scales this ring is detected in spring (CAMBRONY, 1983; KOUTRAKIS & SINIS, 1994). The length overlap between successive age classes is expected for species with an extended reproductive period. The reproduction period of leaping mullet in western Greece occurs from May to November (KATSELIS et al., 1994). This fact reduces the ability to use the 153 length frequency analysis for age estimation (BAGENAL & TESCH, 1978). However, the population of leaping mullet fished in the Messolonghi-Etoliko lagoon system consists of individuals of different ages, which inhabit several biotopes with various levels of trophic importance for fish in adjacent areas of the Ionian Sea and the Gulf of Patras and which enter the lagoons each spring. Thus, a high variability in length at age is expected due to several growth rates. This mechanism could partially explain the fact that between the subareas of the lagoon system there were no differences in lengths at age, the variability of length at age among the years and the length overlap between successive age classes. There are six age classes of leaping mullet in the Messolonghi – Etoliko lagoons. The oldest specimens observed in Mediterranean lagoons (Porto Lagos lagoon: N. Greece) attained an age of eight years (KOUTRAKIS & SINIS, 1994), while maximum age estimates in coastal waters of western Greece was 5 years (CONIDES et al. 1992) and in other localities ranges from 3-5 years. The von Bertalanffy equation obtained in this study appeared to be an accurate description of growth in length of leaping mullet for all stages of its life (fry, juveniles and adults). In our computation the asymptotic total length of the leaping mullet was L∞=32.99±1.25 cm and k=0.258±0.017 year-1. From other Liza species in the Messolonghi –Etoliko lagoons the leaping mullet was estimated to have a smaller L∞ and greater k value. For L. aurata in the same region the von Bertalanffy values have been estimated L∞=69.6 cm, k=0.136 year-1 (HOTOS, 1999), while for L. ramada L∞=56.3 cm, k=0.179 year-1 (MINOS, 1996). These findings were in agreement with those in Porto Lagos lagoon (KOUTRAKIS & SINIS, 1994). Table 4 shows the results obtained by other studies concerning the age determination of leaping mullet. Most of the values of L∞ and k for leaping mullet that inhabit various Mediterranean localities fluctuate between 30 154 and 39 cm and 0.23 and 0.30 year-1 respectively. The majority of the extreme values of L∞ and k are due to the method for the estimation of the length at age or the sample quality. For the leaping mullet inhabiting the Loutsa lagoon (Epirus: NW Greece), GIATNISI (1985) estimates L∞ equal to 42.3 cm and k equal to 0.16 year-1 an implication of the overestimation of L 3 due to small size of the samples (3 specimens). For this reason as well, there are extreme values of von Bertalanffy equation parameters in the SE Mediterranean where the older ages (L4, L5) were demonstrated by 1-2 specimens (EL ZARKA & EL SEDFY, 1970; EL ZARKA et al., 1970), or the overestimation of the lengths at age in older ages due to the length frequency analysis used (DRAKE et al., 1984a; CARDONA, 1999a). From the results shown above it is clear that the growth rates of leaping mullet from different localities can be compared only for young stages and only in some cases (e.g. the debatable results discussed above should not be accepted). Thus, the growth rate of the young leaping mullet (1 to 3 years) in the Messolonghi -Etoliko lagoons is lower than the rates recorded in the other regions of the Mediterranean coastal areas and lagoons (the lagoons of the Gulf of Lions, north and eastern Aegean and the Egyptian coasts). Similar growth patterns of leaping mullet were observed in the Black Sea and finally, the growth of leaping mullet is slightly lower in the Lakes Edku and Quarun in Egypt than in the Messolonghi - Etoliko lagoon (Table 5). Despite the above, we should expect the water temperature to affect growth in length (or weight) patterns of leaping mullet related to latitude because this species does not grow at water temperatures lower than 20ÆC (CARDONA, 1999a). The different growth rates of leaping mullet in different locations can be assigned to other factors apart from temperature differences, such as local food availability and/or density-dependent relations (EL ZARKA et al., 1970; DRAKE et al., 1984b). Feeding of mullet is connected to Medit. Mar. Sci., 3/2, 2002, 147-158 Table 4 Average back calculated lengths (cm) for each year at age and estimate of von Bertalanffy growth equation parameters (± standard error) for year of sampling and all sampling of leaping mullet from back calculated length at age data set. Region W. Greece lagoons Messolonghi Etoliko Loutsa lagoon (Epirus) Porto Lagos lagoon (N. Greece) Lake Vistonida W. Mediterranean (France: Lions Gulf) Atlantic coast (Spain) Balearic Islands Method Ot, vIM Ot, vIM Ot, vIM Sc Sc, vIM Sc, vIM Sc, vIM Sc, vIM Sc Sc Sc Pt Sex M F B B M F B B M F B B L1 10.5 10.4 10.5 9.9 11.5 11.4 11.4 11.4 8.1 8.1 9.2 6.8 L2 15.77 15.8 15.6 14.4 17.8 18 17.9 17.9 13.8 13.8 16.5 18.3 SE Mediterranean (Egypt) Pt Pt Sc M 10.2 16.1 F 10.2 16.1 B 7.6 18.7 Lake Quarun (Egypt) Sc M 12 16.5 Lake Edku (Egypt) Sc Sc F M 12 8.2 16.5 13.5 Sc Ot Ot Ot ? F 9 14.5 M 10.2 15.7 F 9.5 15.5 B 9.8 15.6 B 8.7 14.5 E. Mediterranean (Turkey: NE Aegean) Black Sea Total length (Li)at age i L3 L4 L5 L6 L7 L8 L∞ k t0 Author 19.49 20.98 25.16 29.9 0.30 -0.42 19.5 22.4 25.4 26.4 32.9 0.25 -0.50 Present study 19.4 22.7 24.6 26.6 32.9 0.26 -0.47 21.1· 22.5 42.3 0.16 -0.82 GIATNISI, (1985) 22.1 25.0 27.4 30.7 32.2 33.8 0.32 -0.34 KOUTRAKIS 22.4 25.8 28.2 30.5 32.1 33 36.7 0.28 -0.40 & SINIS, (1994) 22.3 25.6 28.1 30.3 32 33 36.7 0.28 -0.34 21.8 25.6 28.1 36.7 0.28 -0.34 18.9 22.8 26.3 45.2 0.17 -0.17 18.8 23.5 27.7 66.2 0.10 -0.27 EZZAT, (1965)** 22.2 26.2 38.4 0.29 -0.62 CAMBRONY, (1983) 23.4 26.7 28.5 29.9 0.66 -0.14 DRAKE et al., (1984a) 18.1 19.3 22.6 34.25 0.26 -0.44 19.3 22.6 26.5 29.1 38.7 0.20 -0.31 CARDONA, (1999a) 23.6 28.0 32.0 36.4 0.44 0.52 ZAKY-RAFAIL, (1968) 20 24 27.5 36.9 0.23 -0.75 EL ZARKA & EL SEDFY, (1970) 22 27 31 53.6 0.14 -0.82 16.7 21.5 0.50 0.05 EL ZARKA et al., (1970) 17.6 21.6 0.57 0.06 19.8 22.9 25.1 26.8 31.59 0.29 -0.41 20.0 23.4 25.7 27.7 29.0 33.0 0.29 -0.38 KAYA et al., (2000) 19.9 23.2 25.6 27.5 28.8 33.1 0.28 -0.38 17.9 20.9 23 25.8 32.00 0.26 -0.26 ALEEV, (1956)# # cited by THOMSON (1966) , **cited by QUIGNARD & FARRUGIO (1981) Table 5 Growth rate of young leaping mullet in different regions. Total length (Li) at age i L1 L2 L3 Increment from 1 to 3 year (cm) Lake Quarun (Egypt) 12 16.5 20 8 Lake Edku (Egypt) 8.2 13.5 16.7 8.5 Messolonghi (W. Greece: Ionean Sea) 10.5 15.6 19.4 8.9 Black Sea 8.7 14.5 17.9 9.2 E. Mediterranean (Turkey: NE Aegean) 9.8 15.6 19.9 10.1 Lake Vistonida (N. Aegean) 11.4 17.9 21.8 10.4 W. Mediterranean (Lions Gulf) 8.1 13.8 18.9 10.8 Porto Lagos lagoon (N. Aegean) 11.4 17.9 22.3 10.9 W. Mediterranean (France: Lions Gulf) 9.2 16.5 22.2 13 SE Mediterranean (Egyptian coastal) 11 18.7 23.6 12.6 Region Author EL ZARKA & EL SEDFY, EL ZARKA et al., (1970) Present study ALEEV, (1956)# KAYA et al., (2000) KOUTRAKIS & SINIS, (1994) EZZAT, (1965)** KOUTRAKIS & SINIS, (1994) CAMBRONY, (1983) ZAKY-RAFAIL, (1968) # cited by THOMSON (1966) , **cited by QUIGNARD and FARRUGIO (1981) Medit. Mar. Sci., 3/2, 2002, 147-158 155 primary productivity because these fish feed mainly on algae and diatoms, though they consume zooplankton and zoobenthos as well (BRUSLE, 1981). Thus, the lower growth rate which was estimated in this study can be attributed to the fact that the greater sea region of the Messolonghi - Etoliko lagoons is characterised by lower primary productivity values than the north and northeastern Aegean seas (DRAKOPOULOS, et al., 2000) and Gulf of Lions (LEFEVRE, et al., 1997). Growth in length was similar for males and females, even though females were predominant in the older age groups (Table 3). Previous workers have found both significant (EL ZARKA & EL SEDFY, 1970) and nonsignificant differences (KOUTRAKIS & SINIS, 1994; KAYA et al., 2000) in the growth rates of leaping mullet. This disagreement has been found for other Mugilidae species but in these cases the females live longer than males and are predominant in the older age groups (QUIGNARD & FARRUGIO, 1981). The small number of males at older ages can be related to the significant differences in length among the sexes. For example, in the first study (EL ZARKA and EL SEDFY, 1970), the L4 and L 5 was estimated from one male, in contrast to other studies (KOUTRAKIS & SINIS, 1994; KAYA et al., 2000) where the number of male specimens of these ages was greater. This hypothesis is supported by findings for other Mugilidae species in the Mediterranean (KOUTRAKIS & SINIS, 1994; MINOS, 1996; HOTOS, 1999). Finally, the slope of the length-weight relationship (b=3.01≈3), shows that leaping mullet in the Messolonghi –Etoliko lagoons presented an isometric growth pattern. References ALBERTINI-BERHAUT, J. 1974. The biology of young Mugilidae Mugil auratus Risso 1810, Mugil capito Risso 1829, and Mugil saliens, Risso 1810. II. Modifications in the diet in relation to size. Aquaculture, 4:13-27. BAGENAL, T.B. & TESCH, F.W. 1978. Age and growth. p.101-136. In: Methods for assessment of fish production in fresh waters, edited by Bagenal, T.B. IBP Handbook No 3, 3 rd edn Blackwell Publications Oxford. BEN TUVIA, A. 1986. Mugilidae. Vol 3 p.11971206. In : Fishes of the North-eastern Atlantic and Mediterranean, edited by Whitebead, P.J.P., Bauchot, M-L., Hureau, J.- C., Nielsen, J and Totronese E., UNESCO, Richard Clay Ltd Bungay UK. BRULHET, J. 1974. Situation et perspectives des pecheries du mulet jaune de Mauritanie. La Peche Maritime, 1159: 702-706. BRUSLE, J. 1981. Food and feeding in grey mullet. p.185-217. In: Aquaculture of grey mullets, edited by Oren, O. H., Cambridge University Press. CAMBRONY, M. 1983. Recrutement et biologie des stades juveniles de mugilidae (PoissonsTeleosteens) dans trois milieux lagunaires du roussillon et du Rousillon et du Narbonnais (Salses-Leucate,lapalme Bourdigou). Thèse Docteur de 3ème cycle. Université de Perpignan. CARLANDER, J.M. 1987. A history of scale and growth studies of North American fresh water fish. p. 3-14. In: Age and Growth of Fish, edited by Summerfelt, R.C. and Hall, G.E , Ames: Iowa State University Press. CARDONA, L. 1999a. Age and growth of leaping grey mullet (Liza saliens, Risso, 1810) in Minorca (Balearic Islands). Scientia Marina, 63(2):93-93. CARDONA, L. 1999b. Seasonal changes in the food quality, diet feeding rhythm and growth rate leaping grey mullet Liza saliens. Aquatic Living Resources, 12(4): 263-270. CONIDES, ∞., K. ANASTASOPOULOU, G. Acknowledgements PHOTIS, KOUSOURIS, T. & DIAPOULIS, A. 1992. Growth of four Mugilidae stocks in lagoons in western Greece . In: 27th European Marine We thank Dr. N. Fragopoulu, MSc K. Biology Symposium, 7th-11tn September 1992, Koukou and MSc V. Tzanatos for their Dublin. comments on the text. DE SILVA, S. 1980. Biology of juvenile grey mullet: A short review. Aquaculture, 19:21-36. 156 Medit. Mar. Sci., 3/2, 2002, 147-158 DIMITRIOU, E., Y. ROGDAKIS, J. LEONADROS, J. & ATHANASOPOULOS, A. 1994. The quantitative and qualitative composition of the fish catches of the Messolonghi - Etoliko lagoon as an index of fishing management. Fishing News, 155: 82-91 (In Greek). DIMITRIOU, E., KATSELIS, G. & KOUTSIKOPOULOS, C. 2000a. Species diversity of the fishery production of the MesolonghiEtoliko lagoons. Proceedings of 6 th Hellenic Symposium on Oceanography & Fisheries 2: 140145 (In Greek with English abstract). DIMITRIOU, E., ATHANASOPOULOS, A., KAPARELIOTIS, A., KATSELIS, G., ROGDAKIS, Y., BETTS, H. & WENDLING, C. 2000b. Recordings of the hydrology and fishery production of one of the fish farms of the Messolongi-Etoliko lagoon. Proceedings of 9th Panhellenic Symposium of Ichthyologists, 109-112. (In Greek with English abstract). DRAKE, P., ARIAS, A.M. & RODRIGUEZ, R.B. 1984a. Biologia de los mugilidos (Osteicththyes, Mugilidae) en los esteros de las salinas de San Fernando (Cadiz). I. Crecimiento en longitud y peso. Investigaciones Pesqueras, 48: 139-155 DRAKE, P., ARIAS, A.M. & CALLEGO, L. 1984b. Biologia de los mugilidos (Osteicththyes, Mugilidae) en los esteros de las salinas de San Fernando (Gadiz). III Hapitos alimentarios y su relacion con la morfometria del aparato digestivo. Investigaciones Pesqueras, 48:337-367. DRAKOPOULOS, P.G., VALAVANIS, V. & GEORGAKARAKOS, S. 2000. Spatial and temporal distribution of chlorophyll-a in the Aegean Sea according to SeaWiFS imagery. Proceedings of 6 th Hellenic Symposium on Oceanography & Fisheries, 1:206-210 (In Greek with English abstract). EL ZARKA, S. & KAMEL, F. 1965. Mullet fry transplantation and its contribution to the fisheries of inland brackish lakes in the United Arab Republic. General Fisheries Council for the Mediterranean, Proceedings and Technical Papers, 8: 209-226. EL ZARKA, S. & EL-SEDFY, H. 1970. The biology and fishery of Mugil saliens (Risso) in the lake Quarum, UAR. Bulletin of the Institute of Oceanography and Fisheries, Cairo, 1: 3-26. EL ZARKA, S., EL MAGHRABY, A.M. & ABDEL-HAMID, K. 1970. Studies on the Medit. Mar. Sci., 3/2, 2002, 147-158 distribution, growth and abundance of migrating fry juveniles of mullet in a brackish coastal lake (Edku) in the United Arab Republic. Studies and Reviews, General Fisheries Council for the Mediterranean, 46:1-19. ERMAN, F. 1959. Observations on the biology of the common grey mullet (Mugil cephalus). General Fisheries Council for the Mediterranean, Proceedings and Technical Papers, 5: 157-169. GIATNISI, M. E. 1985. Comparative study of biological parameters of genus Mugil (family: Mugilidae). MSc Thesis, Univ. Athens, 105 pp (In Greek). HOTOS, G. 1999. Biology and population dynamics of the golden grey mullet Liza aurata (Risso, 1810) (Pisces:Mugilidae) in the lagoon of MessolonghiEtoliko. Thesis, Univ. Patras, 415 pp (In Greeks). KAYA, M., BILECENOGLU, M. & OZAYDIN, O. 2000. Growth characteristics of the leaping grey mullet (Liza saliens, Risso, 1810), in Homa Lagoon, Aegean Sea. Israeli Journal Aquaculure.Bamidgeh, 52(4): 159-166. KATAVIC, I. 1980. Temporal distribution of young mugilids (Mugilidae) in the coastal waters of the central eastern Adriatic. Acta Adriatica 21(2): 137150. KATSELIS G., MINOS, G., MARMAGAS, A., HOTOS, G. & ONDRIAS, I. 1994. Seasonal distribution of Mugilidae fry and juveniles in Messolonghi coastal waters, western Greece. Bios (Macedonia, Greece), 2:101-108. KENNEDY, M. & FITZMAURICE, P. 1969. Age and growth of thick-lipped grey mullet Crenimugil labrosus in Irish waters. Journal of Marine Biological Association. UK, 49: 683-699 KOTSONIAS, G. 1984. The Messolonghi-Etolico lagoon of Greece. Socio-Economic and ecological interactions of cooperative and independent fishermen. Studies and Reviews, General Fisheries Council for the Mediterranean, 61(2): 521-528. KOUTRAKIS, E. T. & SINIS, A.I. 1994. Growth analysis of grey mullet (Pisces, Mugilidae) as related to age and site. Israel Journal of Zoology, 40:37-53. KOUTRAKIS, E.T, SINIS, A. I. & ECONOMIDIS, P.S. 1994. Seasonal occurrence, abundance and size distribution of gray mullet fry (Pisces, Mugilidae), in the Porto-Lagos lagoon and Lake Vistonis (Aegean Sea, Greece). Israeli Journal Aquaculture-Bamidgeh, 46(4):182-196. 157 LEFEVRE, D., MINAS, H.J., MINAS, M., ROBINSON, C., WILLIAMS, P.J. LE -B. & WOODWARD, E.M.S 1997. Review of gross community production, primary production, net production and dark community respiration in the Gulf of Lions. Deep-Sea Research II, 44 (3-4): 801832. MICKOVIC, B., NIKCENIC, M., HEGEDIS, A. & DAMJANIVIC, I. 1994. Seasonal dynamics of fish fry populations in the brackish waters of the Mrcevo valley. Bios (Macedonia, Greece), 2:143147. MINOS, G. 1996. Biology and population dynamics of the thin-lipped grey mullet Liza ramada (Risso, 1810) (Pisces:Mugilidae) in the MessolonghiEtoliko lagoon. Thesis, Univ. Patras, 272 pp. (in Greeks). MINOS, G., KATSELIS, G., KASPIRIS, P. & ONDRIAS, I. 1994. The differential increase of morphometrical characters during the growth of grey mullet, Liza ramada (Risso) and Liza saliens (Risso), in the Messolonghi - Etoliko lagoon. Bios (Macedonia, Greece), 2:149-154. MINOS, G., KATSELIS, G., KASPIRIS, P. & ONDRIAS, I. 1995. Comparison of the change in morphological pattern during the growth in length of the grey mullet Liza ramada and Liza saliens from western Greece. Fisheries Research, 23:143155. MINOS, G., KATSELIS, G., ONDRIAS, I. & HARRISON, I. J. 2002. Use of melanophore patterns on the ventral side of the head to identify fry of grey mullet (Teleostei:Mugilidae). Israeli Journal Aquaculture –Bamidgeh, 54 (1):12-26. 158 PERLMUTTER, A., BOGRAD, L. & PRUGININ, J. 1957. Use of the estuarine and sea fish of the family Mugilidae (grey mullet) for pond culture in Israel. General Fisheries Council for the Mediterranean, Proceedings and Technical Papers, 4:289-304. QUIGNARD, J. P. & FARRUGIO, H. 1981. Age and growth of grey mullet. p.155-184 In: Aquaculture of grey mullets, edited by Oren, O. H., Cambridge University Press. THOMSON, J. M. 1966. The grey mullets. Oceanography and Marine Biology, Annual Reviews, 4:301-335. THONG, L. H. 1969. Contribution a l'étude de la biologie des Mugilides (Poissons, Teleosteens) des côtes du Massif Armoricain. Bulletin of the Society of Science, Bretagne, 44:137-138. TREWAVAS, E. 1979. Mugilidae. In: Check-list of the fishes of the north-eastern Atlantic and of the Mediterranean, by edited Hureau, J. C and Monod, T , Clofnam I. UNESCO. TORRICELLI, P., TONGIORGI, P. & GANDOLFI, G. 1988. Feeding habits of mullet fry in the Arno River (Tyrrhenian coast). I. Daily feeding cycle. Bollettino di ZooIogia, 3: 161-169. TOSI, L. & TORRICELLI, P. 1988. Feeding habits of mullet fry in the Arno River (Tyrrhenian coast). II. The diet. Bollettino di ZooIogia, 3:171-177. ZAKY-RAFAIL, S. 1968. Investigations of mullet fisheries by beach seine on the UAR Mediterranean coast. Studies and Reviews, General Fisheries Council for the Mediterranean, 3:1-19. ZAR, J.,H. 1984. Biostatistical Analysis. 2nd ed. Prentine-Hall, Englewood Cliffs, NJ, 718 pp. Medit. Mar. Sci., 3/2, 2002, 147-158