Growth studies on Norway lobster, Nephrops norvegicus (L.), in different areas of the Mediterranean Sea and the adjacent Atlantic

(1)

*Received 18 January 1997. Accepted 7 December 1997.

SCI. MAR., 62 (Supl. 1): 43-60

S

_CIENTIA

M

_ARINA

1998

NEPHROPS NORVEGICUS: COMPARATIVE BIOLOGY AND FISHERY IN THE MEDITERRANEAN SEA. F. SARDÀ (ed.)

Growth studies on Norway lobster, Nephrops

norvegi-cus (L.), in different areas of the Mediterranean Sea

and the adjacent Atlantic*

CHRYSSI MYTILINEOU1_{, MARGARIDA CASTRO}2_{, PAULA GANCHO}2

and ANNA FOURTOUNI1

1_{NCMR, Institute of Marine Biological Resources, 16604 Hellenikon, Athens, Greece.} 2_{UCTRA, Universidade do Algarve, Gambelas, 8000 Faro, Portugal.}

SUMMARY: A comparative study of the growth of Nephrops norvegicus among different areas in the Mediterranean Sea and the adjacent Atlantic was conducted. MIX and Bhattacharya’s length-based methods were used for age determination. Both methods were used for all the studied areas. For the estimation of the growth parameters two non-linear methods, based on the results of the length frequency analysis, were used; the Gauss-Newton method, implemented by the SAS pro-gram, was applied using the results of the MIX and the FISHPARM program using the results of the Bhattacharya’s method. The identification of the age groups and their mean lengths-at-age as well as the estimation of the growth para-meters proved to be difficult. A question regarding the adequacy of the von Bertalanffy model was also posed. Remarkable differences were obvious between sexes in the number of identified age groups and their mean lengths-at-age as well as in their growth parameters in all areas. The comparison of the results obtained for the studied areas showed differences, which could not be considered very important except in the case of the Nephrops population of the Alboran Sea, which was characterised by a high growth rate. All other areas seemed to be close; among them the populations from Euboikos Gulf and Catalan Sea being the most different.

Key words: Nephrops norvegicus, crustacean, growth, age, length frequency analysis, Mediterranean, Atlantic.

RESUMEN: ESTUDIO DEL CRECIMIENTO DENEPHROPS NORVEGICUS(L.) EN DIFERENTES ÁREAS DEL MEDITERRÁNEO Y DEL ATLÁNTICO ADYACENTE. – Se realiza el estudio comparativo del crecimiento de Nephrops norvegicus en diferentes áreas del

Mediterráneo y del Atlántico adyacente. Para la determinación de la edad los datos se han analizado utilizando el progra-ma MIX y el método de Bhattacharya. Además, se han utilizado dos métodos no-lineales para la estiprogra-mación de los pará-metros de crecimiento basados en los resultados del análisis de frecuencias de tallas: el método de Gauss-Newton, aplica-do a través del programa SAS y FISHPARM. Para el primer métoaplica-do se utilizaron los resultaaplica-dos de MIX y para el segun-do los resultasegun-dos del métosegun-do de Battacharya. Se notó que la identificación de los grupos de edad y de las tallas medias por edad, así como la estimación de los parámetros de crecimiento, es difícil. Además, se discute la conveniencia del modelo von Bertalanffy aplicado a N. norvegicus. Se han encontrado diferencias en todas las áreas de estudio entre los dos sexos, tanto en el número de grupos de edad identificados y sus tallas medias, como en los valores de los parámetros de creci-miento. Sin embargo, la comparación de los resultados obtenidos para cada área de estudio ha mostrado diferencias que no se pueden considerar importantes, excepto en el caso de la población del Mar de Alborán, la cual indica un ritmo de crecimiento más rápido. Todas las demás áreas tienen crecimientos similares, diferenciándose algo más las del Golfo de Euboikos y las del Mar Catalán.

Palabras clave: Nephrops norvegicus, crustáceos, crecimiento, edad, análisis de frecuencia de tallas, Mediterráneo,

(2)

INTRODUCTION

The age structure of a population and the growth parameters that characterise each stock constitute basic information for stock assessment. The estima-tion of these parameters in crustaceans is difficult due to the absence of permanent hard parts, where age can be registered. In such cases, length frequency analysis has often been chosen to estimate mean lengths-at-age and growth parameters (e.g. Farmer, 1973; Conan, 1978; Hillis, 1979; Sardà, 1985). Many approaches exist for the separation of groups in a length distribution. MIX (MacDonald and Green, 1988) and Bhattacharya’s (1967) method, applied in this present study, are two widely used techniques that have proved useful in estimating mean lengths of the groups present in length distributions of Nephrops (Charuau, 1975; Nicholson, 1979; Figueiredo, 1984; Tully et al., 1989; Mytilineou et al., 1993; Castro, 1995; Mytilineou and Sardá, 1995).

The objective of the present work was a compar-ative study of growth of N. norvegicus among differ-ent areas of the Mediterranean and adjacdiffer-ent Atlantic, using the same length-based methods for all the data. It was assumed that any bias that could possibly be introduced by the length frequency analysis tech-niques used, and arbitrary decisions that such approaches require, will affect all the samples equal-ly, and therefore, the comparative aspects of the study will be valid. Sampling strategies were kept as similar as possible among the different areas.

MATERIALS AND METHODS Sampling strategy

The data consisted of monthly length frequencies for a period of two years, from October 1993 to September 1995 or November 1995 in some areas. The areas sampled were the south coast of Portugal off the Port of Faro in the Atlantic (P), the Alboran Sea off Malaga (M), the Catalan Sea off Barcelona (C), the Ligurian Sea off Genoa (L), the Tyrrhenian Sea off P.S. Stefano (T), the Adriatic Sea off Ancona (A) and the Euboikos Gulf off Athens (G). In some areas such as Portugal, Malaga, Barcelona and the Adriatic the sampling was kept within a small area in the same fishing grounds and data were obtained in a single fishing trip or within a few days in each month. In the other areas, the monthly samples inte-grated data from wider areas.

All samples were obtained by commercial bottom trawlers (except in the Adriatic where a research boat was used) and the mesh of the codend was 40 mm except for Greece (32 mm) and Portugal (55 mm). In all cases the length of the carapace (CL) was measured to the mm below with digital callipers and the data for males, females and ovigerous females were registered separately. Biological samples were also obtained and one of the aspects studied included the number of soft individuals or individuals with gastroliths as well as the number of mature female, information that would allow an estima-tion of the moulting and reproducestima-tion season for each area. This information was used in some aspects of the length frequency analysis dis-cussed further.

Length frequency analysis

Two independent approaches, the MIX (MacDonald and Green, 1988) and Bhatta-charya’s (1967) method, were used by two inde-pendent teams. In each case the objective was to estimate mean lengths for the age groups present in the length distributions by sex, month and area. Data were used in a different way by each team. Team 1, using Bhattacharya’s method implemented in the package FiSAT, entered the data directly as collected, in 1 mm length classes. Team 2, using the program MIX transformed the length frequencies using moving averages of 3 classes on the basic 1 mm length class distribu-tions. This was done to reduce some of the noise introduced by sampling procedures.

Bhattacharya’s (1967) method implemented by the Program FiSAT (Gayanilo et al., 1996) is a method for separating normal components of a distribution with estimation of the mean, stan-dard deviation, separation index (SI; Gayanilo et

al., 1988) and proportions of each one of the

components. A valid application of the method presumes some conventions (e.g. SI ≥2, low

stan-dard deviations, low X2_{values, regressions}

creat-ed as describcreat-ed by Pauly and Caddy, 1985, etc.). In some cases, these were not accomplished at

the same time (especially for X2_{). However, this}

was not considered important enough to reject the analysis, if the identified mean lengths-at-age corresponded to the modes of the distributions

(and since X2_{is meaningless with degrees of}

(3)

MIX (MacDonald and Green, 1988) is a pro-gram that uses a combination of maximum likeli-hood and non-linear estimation methods for esti-mation of the mean, standard deviation and propor-tion of each one of the components of a distribu-tion. In this case, it was assumed that each compo-nent was normal. A criterion was established to select the mean lengths-at-age. They were consid-ered to be the ones visually corresponding to the modes in the distributions, and with standard devi-ations of magnitude not greater than 3 mm. This excluded mean lengths with large standard devia-tions, attributed to mixtures of more than one age group and which were not possible to separate with length frequency analysis.

Once the well represented groups of each method were detected, a relative age in months was attributed to each mean, guided by the following principles: (a) Components identified in the same sample were considered to be at least one year apart. This assumption derives from the fact that this species has a single recruitment period per year. (b) The mean length-at-age of around 15 to 18 mm, showing up for the first time in the winter-spring, was considered to be age 1. For all the areas, the birthday was considered to be 1st February of the year before. The month was indi-cated by the biological data. (c) Growth was con-sidered to be of the type modelled by the von Bertalanffy growth curve. Therefore the increment between two consecutive mean lengths-at-age in the same sample should decrease from smaller to larger sizes. This assumption allowed some of the modes identified in the same sample to be discard-ed or considerdiscard-ed more than one year apart. (d) The time of increase in length should coincide with the time of moulting. Hence, information on moulting provided from the biological data was taken into consideration for interpreting the progression of mean lengths over time. For mature females, infor-mation from the biological data was also taken into account, assuming that ovigerous females cannot moult. (e) The mean lengths-at-age for the same year class should increase with time in order to fol-low as clearly as possible the process of growth by means of a modal progression. That means that a mode representing a year class should shift, after one or more moulting (depending on the sex and the maturity state) to a length representing the next mode to the right (assumed to be one year more). For each area, sex and month, a matrix of values of age and length was obtained per method.

Estimation of growth parameters

Growth parameters of the von Bertalanffy (1957) curve were estimated by using non-linear fit-ting procedures. In one case, when MIX was applied to estimate mean lengths-at-age, the Gauss-Newton method implemented by the program Statistical Analysis System (SAS Institute Inc., 1989) was used. In the other case, when Bhattacharya’s method was applied for estimation of mean lengths-at-age, the FISHPARM program implemented by the soft-ware package FSAS (Saila et al., 1988) was used. Both approaches to non-linear estimation operate under similar assumptions. In the present study, the analysis was based on the month’s data, and not on those of year classes, since modal progression analysis was not considered very useful for growth estimation (Castro et al., 1998).

The growth parameters were estimated by the two methods for each area, sex and month separate-ly. 1) In many cases, the growth parameters estimat-ed by the Gauss-Newton method and SAS program were totally unreasonable. Sometimes extremely

high values of L∞and low values of k were obtained.

A criterion for acceptance of values was therefore established; only months that fitted the model were accepted. Parameter estimates that produced values

of L∞larger than 80 mm for females or 100 mm for

males were also dropped. 2) For many months, the results obtained by the FISHPARM program were

found unreasonable (L∞ lower than the maximum

observed length or higher than any value mentioned in published literature - higher than 100 mm for males and 80 mm for females). All the months with unreasonable growth parameter estimates were dis-carded. Some more criteria were also established; the presence of the younger and older mean lengths-at-age and, the increments between the mean lengths-at-age to be decreasing with age. Once the months with samples containing information for parameter estimation were selected, a global esti-mate for each area and sex was obtained per method. Moreover, since the estimation of the growth parameters is related to the quality of the analysed sample and the quality of the identified mean lengths-at-age, one more attempt to estimate growth parameters was done using the Bhattacharya’s method results, but with more con-ventions apart from the above mentioned; only sam-ples with a high number of specimens without many gaps in their length composition and absence of gaps between the identified mean lengths-at-age

(4)

were used. In this case, only one month producing acceptable estimates and accomplishing, if possible, almost all the above criteria, was used. Finally, the estimation of growth parameters was also attempted using all the identified mean lengths-at-age per studied area for comparison purposes.

Comparison of the data and results

In order to examine the closeness of the studied areas with respect to the mean lengths of the month-ly samples, a cluster anamonth-lysis was applied. Because of the different selectivity of the gears used, only lengths more than 30 mm CL were included (the length of 30 mm is greater than L50 of the selection curves of the gears used; Sardà et al., 1993; Mytilineou et al., 1998).

Cluster analysis was also used to examine dis-tances between the studied areas with respect to the estimated mean lengths-at-age in the various months. Age 4, one of the best represented in the samples, was selected for the analysis, because it had the highest number of months with data avail-able for all seven areas. This analysis was applied to the results of Bhattacharya’s method. In addition, pairwise comparisons of the mean lengths-at-age between areas were carried out by determining the absolute difference between the mean lengths-at-age of two areas, expressed as a percentlengths-at-age of their

mean (APD: (|X1- X2|) 100 / (X1+ X2))/2, where X1:

mean length-at-age of the area 1 and X2: mean

length-at-age of the area 2). The mean absolute per-centage difference (MAPD: mean of all individual APD values for each area) was also used to compare each area as a whole.

Pairwise comparisons of the growth parameters obtained by the Gauss-Newton method for each area

and sex were made using Hotelling’s T2_{test for}

glob-ally comparing the sets of estimated parameters without assuming equal variance-covariance matri-ces (Bernard, 1981; Hanumara and Hoenig, 1987). The matrix algebra language IML (SAS Institute Inc., 1989) was used to perform the calculations, using as base data the results of the non-linear pro-cedure PROC NLIN (SAS Institute Inc., 1989) as described by Hanumara and Hoenig (1987).

The growth performance index φ (Pauly and Murno, 1984) was also estimated for comparison purposes from the growth parameters derived by the FISHPARM program (analysis of selected months). The calculation was made according to the

equa-tion: φ = log10k + 2log10 L∞.

Other methodologies

Along with this work, many variations of the described methodologies were tried and abandoned. These included different grouping of the base data, estimation of growth parameters based on the incre-ments of each cohort followed over time, estimation

of growth parameters forcing L∞over given values and

the use of the Ford-Walford method (Walford, 1946). When a methodology led to unreasonable results, it was abandoned. The original length data were also used for the direct estimation of growth parameters, applying ELEFAN I (Pauly and David, 1981; Gayanilo et al., 1988) and Shepherd’s (1987) method implemented by the FISAT package (Gayanilo et al., 1996). In most cases, k was found to be very high, as

was L∞too. The option of choosing or forcing L∞to

more reasonable values resulted in bad fitting curves. For this reason, these two methods were also aban-doned.

RESULTS

The monthly length frequency distributions of

Nephrops norvegicus for the two years of sampling

were examined for the various studied areas separate-ly for each sex (distributions not shown here). Table 1 presents information related to the minimum and maximum CL of Nephrops specimens caught in the different areas during the present study. The examina-tion of the distance between the monthly mean lengths in the various areas showed that for males A, T, L, P and G were close, while C constituted a different clus-ter and M was even more isolated. For females, A, T, L, P and C were very close, while G a little further apart. M again constituted a very different cluster. The results of this cluster analysis are shown in Figure 1.

TABLE 1. – Minimum and maximum carapace length (mm) of

Nephrops specimens caught during the two years of sampling in each

area for each sex separately.

Males Females

AREA Min. Max. Min. Max.

Atlantic 11 63 10 50 Alboran Sea 14 60 13 48 Catalan Sea 11 79 10 66 Ligurian Sea 12 63 10 55 Tyrrhenian Sea 16 75 17 60 Adriatic Sea 20 65 19 54 Euboikos Gulf 10 63 11 52

(5)

In the length frequencies, certain modes appeared more consistently than others. In particu-lar, modes relating to lengths between 25 mm and 40 mm CL, appeared in most samples from all areas. These modes presented a continuity over time, indicating that they represented modes relat-ed to different year classes and not to moulting activity. However, in some samples the modes were very poorly detected. In Figure 2, two examples, a “good” and a “poor” sample, are shown and in Figure 3 some examples of the continuity of modes over time are presented.

Length frequency analysis was found to be quite difficult and complicated. From all the analyses, it was obvious that the main problem in the length-based analysis was not the selected method, but both the sample structure and the way of applying the analysis (criteria used). All values of the esti-mated mean lengths-at-age resulting from the two procedures are presented in Tables 2, 3, 4 and 5. The visual comparison of the results of the two pro-cedures (MIX and Bhattacharya’s) showed that, in

most of the cases, they were very close. An exam-ple of the mean lengths-at-age estimated by the two approaches is presented in Figure 4.

According to Bhattacharya’s method (Tables 2 and 4), the maximum identified number of age groups during the two years of study was for males: 9 for P, 9 for M, 8 for C, 10 for L, 9 for T, 8 for A and 10 for G and for females: 6 for P, 8 for M, 6 for C, 7 for L, 8 for T, 7 for A and 9 for G. In most cases, it was difficult to detect the youngest (0+ and 1) and oldest (>5 for males and >4 for females) age groups. According to MIX (Tables 3 and 5), the maximum identified number of age groups during the two years of study was for males: 7 for P, 8 for M, 8 for C, 8 for L, 9 for T, 8 for A and 11 for G and for females: 6 for P, 8 for M, 7 for C, 7 for L, 6 for T, 7 for A and 8 for G. Again, it was difficult to detect the youngest (0+ and 1) and oldest (>5 for males and >4 for females) age groups in most cases. In both analyses, it was obvious that more age groups occurred in the length frequencies of all areas, but they could not be identified because of their low representation in the samples. In all areas two or three cohorts were well represented and could be followed along most of the samples.

25

20

15

10

5

0 _{A T LPGCM}

_{AT LPC GM}

MALES

FEMALES

A

verage Linkage

FIG. 1. – Cluster analysis using the Average Linkage method for the examination of the distances between the mean lengths (CL) of the monthly samples of the various areas for each sex. Faro in the Atlantic (P), the Alboran Sea off Malaga (M), the Catalan Sea off Barcelona (C), the Ligurian Sea off Genoa (L), the Tyrrhenian Sea off P.S. Stefano (T), the Adriatic Sea off Ancona (A) and the

Euboikos Gulf off Athens (G)

0 20 40 60 Catalan-Males May 1994 N=54 N of individuals 10 8 6 4 2 0 10 20 30 40 50 60 70 Portugal-Females December 1993 N=24 CL (mm) 10 20 30 40 50 60 CL (mm)

FIG. 2. – Examples of a “good (Catalan-males) and a “poor” (Portugal-females) length frequency distribution. In the first distri-bution the modes are clearly observed in contrast to the second one.

(6)

These corresponded well with the two or three modes observed in the length frequencies, reported above. This fact constituted a confirmation for the validity of the mean lengths-at age identified by the length frequency analysis.

The results of the length frequency analyses indicated that differences exist between the sexes in all the areas. The females always presented fewer age groups and lower increments than the males (Table 2, 3, 4 and 5). Regarding the differences

Mean length-at-age 15 20 25 30 35 40 45 50

Oct 93 Nov 93 Dec 93 Jan 94 Feb 94 Mar 94 Apr 94 _{May 94} Jun 94 Jul 94 Aug 94 Sep 94 Oct 94 Nov 94 Dec 94 Jan 95 Feb 95 Mar 95 Apr 95 _{May 95} Jun 95 Jul 95 Aug 95 Sep 95 Oct 95 Nov 95

Tyrrhenian-Males 15 20 25 30 35 40 45 50

Tyrrhenian-Females Year Class A Year Class B Year Class C MOLT 15 20 25 30 35 40 45 50

Months Malaga-Males 15 20 25 30 35 40 45 50

Malaga-Females

FIG. 3. – Examples of the continuity of modes over time for three year classes (the most representative in the samples). Increase in length is observed after moulting, and is more easily detected in adult females (restricted moulting period) than in males (moulting all the year).

(7)

93 93 93 94 94 94 94 94 94 94 94 94 Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep A TLANTIC 30.91 28.00 29.00 27.91 34.50 28.00 24.50 24.00 27.61 28.00 24.33 37.31 33.34 34.41 36.00 39.79 33.89 30.44 31.90 33.34 33.87 29.49 42.99 37.89 39.54 43.20 44.71 38.14 36.21 37.97 38.93 40.10 35.29 51.26 43.52 44.37 48.77 48.57 43.67 41.46 42.75 44.58 44.21 40.05 47.69 48.71 51.69 47.21 47.95 49.81 48.99 44.16 52.67 55.67 52.71 52.67 53.27 49.74 59.67 54.34 N 586 37 262 73 131 161 290 229 795 136 377 ALBORAN 34.30 36.03 32.16 41.25 28.50 27.26 37.67 30.30 30.90 29.28 31.33 31.77 41.80 41.52 37.39 47.07 36.41 36.36 43.74 37.13 40.96 38.38 38.28 38.43 45.88 47.29 42.57 52.71 42.71 41.41 49.08 43.82 45.45 44.00 44.65 44.24 51.08 51.89 47.29 58.45 48.44 48.13 54.32 49.47 48.85 49.05 49.65 49.08 55.26 55.38 51.51 63.67 53.45 52.73 58.16 54.75 53.77 53.75 54.76 53.67 59.24 56.73 68.37 58.55 58.12 61.70 58.46 57.17 57.36 59.18 61.67 59.26 61.32 62.77 60.67 63.14 65.09 N 736 198 399 107 414 549 389 333 233 420 202 198 CA T ALAN 27.75 22.98 21.99 23.33 19.55 20.39 21.48 23.70 22.71 22.03 27.59 22.23 32.15 28.94 28.08 28.23 26.12 26.40 30.14 30.80 31.71 28.91 35.15 28.23 37.87 34.73 33.22 34.10 32.16 32.28 35.98 36.84 37.10 36.00 40.31 33.37 43.51 39.16 37.92 39.44 36.78 37.78 41.70 40.17 41.68 42.48 45.57 38.44 44.19 43.61 44.66 41.76 42.47 46.47 45.03 46.35 49.05 42.84 48.48 51.31 49.35 46.40 50.53 48.23 51.62 53.98 54.10 N 106 880 681 365 563 194 850 544 919 381 307 503 LIGURIAN 16.19 24.00 29.33 20.00 27.00 29.75 23.13 30.51 36.27 33.50 33.30 39.14 29.29 35.71 41.50 39.41 39.35 46.35 35.51 41.26 46.00 44.34 45.42 50.69 41.51 46.87 50.13 48.88 49.28 47.01 50.79 54.37 53.67 54.79 58.90 58.07 N 251 553 503 97 337 193 TYRRHENIAN 20.70 17.00 23.89 25.39 23.20 18.28 19.99 18.40 16.00 18.65 16.50 23.17 28.18 22.50 29.09 31.31 30.19 24.25 26.27 27.00 21.73 27.33 22.50 30.00 34.90 28.34 36.30 35.83 35.26 31.70 31.72 33.07 26.50 34.01 29.92 35.61 41.54 36.07 41.35 42.83 40.63 39.65 38.15 38.54 33.24 40.56 35.57 40.1 1 47.94 40.22 45.57 47.38 45.52 44.86 44.49 44.24 38.25 45.61 40.62 44.80 53.48 45.90 52.50 50.29 49.82 49.76 47.69 42.77 49.15 46.32 49.17 57.32 57.03 54.95 54.78 55.92 51.33 47.25 52.83 50.68 52.49 51.04 N 480 881 737 579 826 595 761 333 327 407 403 306 ADRIA TIC 22.98 25.84 26.57 16.50 15.94 17.33 19.79 20.21 15.00 17.71 23.67 15.67 29.71 31.13 33.78 21.70 22.89 25.16 27.01 27.40 22.76 23.04 29.57 26.64 36.27 37.72 38.83 26.84 26.99 31.01 33.95 33.47 29.81 30.97 34.57 35.07 43.22 43.85 43.53 33.20 34.77 36.16 39.57 39.38 35.37 36.47 39.66 40.95 47.56 47.32 47.92 39.20 39.40 40.88 44.91 44.75 40.75 40.91 43.42 46.65 51.17 43.60 44.13 45.66 49.02 49.30 45.40 45.03 48.85 48.92 49.52 53.1 1 49.33 N 1606 835 630 565 487 202 879 653 788 491 350 853 EUBOIKOS GULF 24.24 24.76 14.00 23.70 18.46 18.12 24.70 23.53 23.42 20.67 20.00 21.67 30.43 31.59 22.54 29.91 25.75 26.64 31.06 33.20 29.91 27.99 26.08 27.55 38.05 38.53 28.25 35.68 31.43 32.97 37.22 39.43 34.62 34.87 33.48 34.87 44.31 44.28 33.87 39.63 36.21 39.42 43.92 44.91 40.15 38.74 39.20 40.49 50.35 49.23 40.08 44.38 40.89 45.09 50.30 51.36 45.79 43.84 44.77 45.81 54.84 53.84 45.80 50.1 1 50.32 51.87 55.92 55.57 51.60 49.63 49.28 53.24 50.57 54.24 58.83 56.16 57.50 54.40 N 2254 2535 1825 1200 806 764 375 347 405 344 726 689 94 94 94 95 95 95 95 95 95 95 95 95 95 95 Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov 31.59 37.00 35.39 36.00 35.51 38.1 1 24.00 27.51 27.52 36.28 41.32 40.67 41.52 42.82 44.84 30.50 30.80 34.45 41.76 45.90 44.83 47.70 49.80 49.59 35.33 35.59 41.84 47.31 50.70 49.09 51.23 55.02 40.21 40.88 46.30 50.85 52.21 45.51 46.22 50.34 54.77 49.93 49.32 58.00 59.76 53.67 54.02 844 55 86 107 99 103 300 321 318 33.38 33.80 34.81 26.32 25.91 27.04 28.76 29.88 30.50 26.33 31.28 33.79 31.00 39.81 39.49 41.90 39.37 35.82 35.12 36.55 36.06 43.50 32.60 37.64 38.87 40.70 45.68 44.94 47.44 47.41 42.42 41.72 43.49 43.20 49.57 37.31 44.35 44.1 1 46.64 50.40 49.98 51.55 52.87 47.96 47.90 48.84 49.54 53.62 49.06 49.03 49.62 50.16 56.10 58.46 52.10 52.77 53.30 54.49 57.51 53.76 54.67 56.46 56.95 56.84 59.47 60.49 60.67 134 11 0 250 315 388 185 284 125 123 539 372 312 430 28.86 25.04 23.60 18.10 20.91 19.76 23.70 24.06 22.88 18.00 29.30 23.48 24.27 25.46 35.65 30.25 30.03 24.65 27.09 27.66 29.12 30.70 29.83 25.49 36.52 29.72 29.81 29.85 39.83 35.10 36.49 30.67 32.06 32.71 32.44 35.30 36.20 31.93 41.80 35.46 35.15 34.17 43.32 39.89 42.29 36.50 37.08 37.10 37.59 39.78 41.04 36.87 47.67 40.51 40.19 39.67 44.43 46.27 42.14 40.36 42.57 41.33 45.43 41.19 45.50 45.14 43.18 49.33 46.67 45.73 46.90 48.50 49.98 47.67 51.47 48.34 52.40 53.69 559 563 472 405 348 361 290 335 471 1897 587 382 730 319 14.00 13.33 16.00 16.42 18.12 20.50 20.31 22.62 22.88 27.26 28.00 21.82 22.99 23.93 24.12 27.47 27.53 28.95 29.38 28.67 33.80 35.57 27.90 30.55 30.35 30.17 33.60 33.54 36.96 35.78 34.20 40.09 42.78 33.86 36.91 36.41 35.75 37.94 38.41 42.02 41.63 39.06 44.33 47.63 39.62 42.64 41.84 41.01 42.00 47.43 46.91 43.30 51.83 44.42 47.49 46.17 45.35 47.17 50.61 51.58 51.68 50.74 49.80 126 490 435 208 171 159 477 1158 561 479 373 22.42 17.00 24.43 24.75 24.67 18.67 21.92 21.95 21.21 21.51 15.67 24.84 18.50 14.00 26.90 24.75 31.00 31.70 32.82 26.04 28.29 28.17 27.90 27.62 21.85 31.92 24.83 24.79 34.71 30.33 36.25 37.27 37.99 32.78 34.85 33.18 34.65 33.81 27.56 37.22 30.99 30.16 41.77 36.51 41.42 43.18 43.43 37.61 43.07 37.24 39.72 39.66 33.49 42.10 36.98 35.60 45.95 41.51 46.50 47.81 47.45 43.28 47.88 41.84 43.99 44.07 38.33 46.69 42.93 40.41 53.39 47.31 50.95 51.59 54.06 49.06 52.26 46.64 47.58 49.27 42.96 50.35 48.34 45.38 51.98 55.00 56.51 51.50 51.33 52.57 54.29 51.65 51.67 1090 574 799 788 384 268 574 631 637 618 1057 572 642 678 15.00 16.80 17.42 20.77 18.52 23.82 18.78 17.50 19.49 23.99 23.64 23.44 27.72 24.73 29.68 24.68 24.80 24.05 29.86 29.18 29.49 34.09 30.61 34.75 30.44 29.05 28.92 34.64 36.14 34.96 39.92 35.96 40.22 35.46 34.46 34.95 39.56 41.40 40.52 44.91 41.30 45.37 39.76 39.08 39.01 44.43 45.32 46.00 49.84 46.14 44.51 44.67 44.80 48.93 50.69 51.84 53.67 50.18 47.86 48.96 53.13 51.60 370 659 275 786 662 497 535 668 780 19.04 13.33 22.67 15.92 16.72 20.81 21.76 20.98 28.33 19.33 28.21 21.50 28.65 21.76 32.43 24.80 25.86 27.48 27.71 27.31 34.64 26.37 33.82 28.37 38.66 29.74 39.06 32.05 31.66 32.82 34.56 33.1 1 40.03 33.31 40.80 35.97 44.96 38.24 45.73 39.46 36.82 38.74 39.99 38.82 44.64 40.86 45.30 40.81 52.60 44.26 51.97 44.60 41.95 46.05 45.66 44.41 52.33 46.17 52.29 45.64 49.33 58.66 50.63 47.56 53.82 51.00 49.33 56.70 52.16 52.07 53.52 56.87 53.62 59.00 56.63 55.86 56.93 58.67 57.36 58.65 58.28 1505 644 402 535 669 297 288 1102 490 406 286 590 T ABLE

2. – Identified mean lengths-at-age of male

N. norvegicus using BHA TT ACHAR Y A

’S method and number (N) of examined individuals from Octobe 1993 to November 1995 in the dif

ferent studied a

(8)

93 93 93 94 94 94 94 94 94 94 94 94 Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep A TLANTIC 30.96 29.96 36.17 27.26 33.76 22.09 25.80 24.36 26.91 27.57 25.26 38.36 38.06 42.89 33.22 39.58 28.70 31.67 31.36 31.46 33.19 29.59 43.63 42.16 59.20 42.77 45.17 35.99 39.25 37.03 37.35 40.20 35.89 49.89 46.65 65.56 47.98 49.39 43.61 45.91 42.31 43.68 47.00 43.35 56.77 53.52 56.50 56.90 49.20 53.16 47.98 50.64 58.20 50.19 65.56 60.55 58.31 58.12 54.57 59.35 57.42 N 586 37 262 73 131 161 290 229 795 136 377 ALBORAN 30.46 28.41 32.85 34.12 33.95 25.61 30.07 27.20 28.90 28.91 24.84 26.20 38.26 35.76 37.65 40.87 43.57 35.94 37.90 30.00 35.49 38.80 30.69 31.96 43.30 41.10 42.03 46.25 52.67 41.26 42.90 33.82 41.19 44.21 37.02 39.01 48.54 50.92 46.34 52.54 59.23 50.21 48.56 41.98 47.00 48.99 43.84 46.1 1 55.35 62.01 49.34 60.49 64.93 58.39 54.61 48.62 53.51 54.87 48.32 53.15 63.73 67.44 53.13 68.05 73.62 63.58 58.70 55.60 57.31 62.31 54.19 60.19 71.04 74.03 59.01 69.19 62.37 60.43 64.76 67.50 60.00 64.55 67.33 N 736 198 399 107 414 549 389 333 233 420 202 198 CA T ALAN 21.1 1 23.50 22.35 23.80 19.67 21.31 21.37 23.68 22.62 22.39 26.82 23.35 27.21 28.14 27.64 28.33 26.10 26.61 30.88 30.76 31.27 29.34 34.07 28.06 31.39 33.16 32.91 33.69 32.22 31.91 41.63 36.46 36.82 36.21 38.74 33.03 37.16 37.81 38.07 38.84 40.09 37.94 46.95 40.53 43.16 43.20 44.73 38.36 42.42 43.46 44.35 44.50 46.51 46.00 53.72 46.39 50.37 48.60 49.00 42.44 48.43 48.19 50.43 49.94 50.34 58.56 54.17 56.24 55.58 58.08 47.27 54.52 53.58 56.76 54.57 N 106 880 681 365 563 194 850 544 919 381 307 503 LIGURIAN 16.23 27.67 36.04 20.45 17.44 21.33 22.59 33.25 41.14 28.28 28.05 29.78 34.38 38.69 46.1 1 33.61 33.17 38.28 46.16 46.18 53.86 38.65 38.33 45.23 57.63 55.16 60.91 45.13 44.29 53.09 53.66 51.82 61.04 59.00 N 251 553 503 97 337 193 TYRRHENIAN 22.58 27.94 22.94 17.34 14.45 19.85 19.42 15.02 15.15 18.25 16.94 17.34 27.95 37.20 28.77 32.79 22.64 31.46 28.98 23.51 21.14 24.63 22.93 26.36 34.60 41.75 34.64 43.38 32.15 40.29 36.98 32.22 26.32 32.99 27.97 32.17 43.15 48.15 39.59 47.85 39.14 45.55 45.17 37.33 34.25 39.70 33.22 39.12 46.39 57.06 44.95 53.26 46.60 50.00 50.59 44.41 41.33 46.77 40.28 44.74 53.56 68.56 53.23 62.51 53.59 55.21 57.55 55.59 46.48 53.78 47.03 46.79 57.79 50.41 62.59 52.14 53.00 56.55 58.89 N 480 881 737 579 826 595 761 333 327 407 403 306 ADRIA TIC 23.83 25.55 17.18 16.26 16.07 17.21 19.91 19.99 22.19 22.86 24.21 15.92 29.91 30.62 26.74 25.62 27.23 25.53 27.41 28.31 29.97 32.55 29.33 26.71 37.32 37.54 33.82 32.18 32.61 31.49 33.93 35.93 37.36 39.37 34.34 34.82 42.50 46.39 40.85 37.61 37.09 36.14 39.63 42.19 42.86 43.85 39.36 40.47 52.24 60.19 47.74 43.96 43.59 40.71 44.62 48.06 49.62 51.25 43.91 47.35 59.31 53.34 49.69 48.53 45.07 53.63 54.03 58.56 52.12 57.23 59.56 58.30 64.02 51.47 60.10 59.56 N 1606 835 630 565 487 202 879 653 788 491 350 853 EUBOIKOS GULF 30.26 30.15 22.23 31.68 25.50 26.48 23.04 33.14 22.91 27.19 25.91 26.86 39.94 37.37 27.54 37.82 30.90 32.93 30.43 39.12 32.00 32.91 33.57 34.27 48.50 45.68 33.40 43.71 37.13 43.68 36.80 45.40 39.84 37.74 38.56 40.12 56.1 1 53.94 40.09 50.19 43.54 54.88 43.68 51.43 45.82 44.03 43.19 46.29 61.79 61.14 46.04 54.54 48.91 63.28 51.80 57.38 51.83 50.07 47.91 54.17 67.62 70.97 52.28 59.27 55.32 68.01 60.57 69.21 57.39 56.21 55.70 61.13 59.34 66.36 62.56 61.66 61.07 65.64 67.44 67.30 N 2254 2535 1825 1200 806 764 375 347 405 344 726 689 94 94 94 95 95 95 95 95 95 95 95 95 95 95 Oct Nov Dec Jan Feb Mar A pr May Jun Jul Aug Sep Oct Nov 32.61 31.94 29.08 29.76 24.43 20.44 26.66 27.66 28.13 37.76 36.54 35.23 35.45 34.88 28.68 31.09 34.66 34.84 44.07 40.68 41.08 41.07 42.69 37.03 35.59 41.14 42.50 51.34 45.06 49.13 47.90 50.00 43.55 39.93 47.80 51.29 50.60 52.87 52.22 54.50 49.16 44.33 53.87 58.49 56.84 60.56 58.65 58.14 58.60 49.79 57.72 64.56 62.56 56.59 844 55 86 107 99 103 300 321 318 28.36 19.58 33.91 25.54 20.89 26.40 25.29 24.12 30.75 29.03 30.96 35.36 19.82 32.78 27.50 38.70 38.79 25.96 39.25 31.32 29.98 36.66 33.14 36.91 40.06 32.1 1 39.98 33.00 43.70 46.63 35.06 47.00 42.50 34.54 42.12 37.40 43.08 45.24 38.92 46.93 37.22 47.90 52.22 40.04 51.87 48.77 40.81 46.10 42.45 48.60 49.46 42.84 56.10 47.64 51.84 58.71 46.14 56.60 53.89 45.37 51.94 48.43 52.33 53.70 50.31 61.53 58.53 57.54 65.22 51.61 62.50 57.44 50.32 56.19 53.82 57.07 58.10 55.12 64.89 56.72 62.1 1 54.72 61.66 63.28 63.97 59.94 60.92 58.14 65.56 67.24 63.68 134 11 0 250 315 388 185 284 125 123 539 372 312 430 23.38 25.12 23.29 17.92 19.82 20.84 20.66 24.1 1 22.04 21.39 24.02 23.60 24.02 27.32 27.49 30.21 27.96 24.97 25.65 27.04 24.96 31.52 29.20 25.76 29.17 29.14 28.77 32.70 31.73 38.01 33.03 30.10 30.81 31.88 31.85 36.02 35.99 31.1 1 35.42 34.62 34.05 41.19 41.79 44.38 41.88 35.24 36.97 36.95 37.22 46.03 41.05 35.92 41.22 40.25 40.75 45.36 53.20 46.47 42.03 40.23 42.06 43.52 47.97 41.01 45.77 43.85 45.13 50.22 49.75 44.52 51.55 50.08 54.16 45.18 53.56 48.98 49.82 55.56 49.03 58.62 50.09 54.13 54.44 559 563 472 405 348 361 290 335 471 1897 587 382 730 319 15.34 13.38 16.17 15.70 18.07 19.35 22.83 23.82 26.14 16.38 20.61 20.94 23.10 23.87 27.46 25.17 29.64 30.66 31.92 23.05 25.18 24.73 28.63 29.12 36.80 29.82 37.61 35.62 40.45 27.64 32.94 30.68 35.41 35.03 46.54 38.29 46.92 41.33 46.06 34.81 39.61 35.57 41.58 40.00 53.76 50.03 52.25 46.62 54.56 43.48 46.05 42.60 46.69 46.26 59.63 63.61 53.56 49.74 54.20 50.50 53.68 51.54 55.34 60.57 58.48 126 490 4350 208 171 159 477 1158 561 479 373 14.23 14.83 27.26 24.50 16.44 19.47 18.48 22.14 20.83 21.81 22.16 24.97 16.98 14.83 27.02 25.23 35.39 30.49 24.66 27.95 23.16 26.91 26.17 31.73 27.08 31.43 25.12 24.1 1 35.45 30.53 42.05 35.30 34.12 32.93 30.12 32.06 32.15 38.31 32.65 36.75 33.54 28.95 42.18 37.08 47.24 43.83 41.79 37.74 35.55 36.82 38.10 44.67 37.62 42.09 40.33 33.91 47.92 42.52 53.71 50.00 47.17 43.99 43.19 46.28 45.14 49.48 45.51 48.07 48.63 39.67 53.21 49.36 57.30 53.09 52.43 50.50 52.93 50.44 53.96 53.23 54.76 55.83 45.32 59.99 54.34 62.56 58.88 57.28 59.56 56.16 50.87 54.94 60.56 1090 574 799 788 384 268 574 631 637 618 1057 572 642 678 14.21 17.06 17.13 20.82 19.67 18.86 17.86 18.39 20.86 25.23 25.85 22.16 27.59 24.33 23.66 23.95 23.42 27.86 30.14 30.45 29.26 33.24 30.07 28.39 29.42 28.32 34.69 39.31 37.25 34.08 39.04 36.15 33.88 35.26 34.17 38.85 46.14 44.83 40.75 43.27 44.39 39.83 39.98 39.44 45.01 59.56 51.49 47.09 49.57 49.48 47.43 46.21 45.48 49.62 59.63 51.46 58.73 55.64 52.68 51.99 51.20 55.92 56.97 56.05 57.58 61.57 370 659 275 786 662 497 535 668 780 38.78 21.62 25.83 17.10 25.59 20.98 20.54 27.72 28.21 25.18 27.87 28.91 44.43 29.09 31.72 22.1 1 31.29 29.14 25.39 32.07 33.87 29.96 35.04 37.10 52.39 36.32 37.32 25.68 35.72 34.60 30.40 37.09 39.64 35.17 41.06 44.67 58.34 49.70 44.03 31.33 40.96 40.02 36.60 41.77 44.09 40.95 48.20 53.19 65.49 57.53 38.96 47.15 46.85 44.74 47.07 48.24 47.07 54.59 59.86 69.65 46.75 56.1 1 55.30 52.73 55.67 52.1 1 52.06 64.55 71.07 55.80 64.35 61.15 59.82 65.91 56.01 56.41 64.21 62.50 64.52 66.44 71.52 1505 644 402 535 669 297 288 1102 490 406 286 590 T ABLE

3. – Identified mean lengths-at-age of male

N. norvegicus

using MIX method and number (N) of examined individuals from October 1993 to November 1995 in the dif

(9)

Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep 93 93 93 94 94 94 94 94 94 94 94 94 A TLANTIC 29.19 27,00 29.63 25.50 29.17 27.00 27.77 25.47 25.47 29.50 30.1 1 33.23 30.34 33.90 30.28 33.95 32.16 33.88 30.35 30.22 34.44 35.31 36.87 39.33 38.73 38.00 39.12 36.18 41.97 34.59 34.55 39.63 41.50 40.47 39.94 39.65 38.37 44.00 44.67 45.69 42.81 45.42 N 256 24 83 44 74 153 328 227 985 64 164 ALBORAN 30.01 34.68 30.06 35.70 33.63 24.50 33.79 28.33 28.00 27.20 30.92 32.57 34.49 38.73 35.00 41.52 38.73 32.72 39.24 35.87 34.57 33.58 39.18 38.53 38.86 39.98 45.90 43.66 39.10 43.81 40.86 39.72 39.35 44.81 44.22 43.22 49.94 47.42 44.39 48.70 45.18 43.90 44.92 52.77 50.00 48.75 48.71 52.44 49.28 48.63 49.48 51.58 52.96 53.24 56.09 N 710 186 202 56 289 650 644 923 601 798 205 191 CA T ALAN 23.71 22.54 22.79 18.50 20.90 19.80 20.09 23.29 22.66 22.81 21.70 23.13 27.07 27.87 27.58 24.98 25.57 26.40 26.01 29.10 31.31 28.81 26.79 27.33 30.81 31.76 33.63 29.94 30.77 31.59 31.75 32.64 36.25 32.71 32.87 31.66 34.22 35.32 37.33 33.78 35.45 36.61 36.07 36.04 36.84 36.90 40.67 40.81 N 9 1 655 526 302 554 281 1196 630 1100 355 237 356 LIGURIAN 16.02 27.67 34.79 33.18 35.98 29.61 21.86 33.73 40.25 39.1 1 40.16 36.33 28.73 39.02 44.33 44.28 44.26 40.68 33.50 40.40 N 240 437 389 87 460 173 TYRRHENIAN 19.97 16.33 24.05 25.14 27.21 18.91 12.00 22.00 17.67 20.60 20.43 20.33 26.87 23.48 29.45 31.50 31.91 26.65 20.19 31.92 24.12 26.57 30.03 27.87 32.01 28.18 34.26 36.03 35.91 31.47 26.56 37.73 29.45 31.18 35.31 30.86 37.44 32.84 38.04 39.52 35.14 32.91 41.91 34.41 35.37 39.73 34.14 36.41 38.39 38.12 40.25 38.58 44.45 42.82 42.62 N 447 712 547 474 761 622 824 241 340 398 439 250 ADRIA TIC 22.82 24.06 25.76 16.44 16.02 16.86 19.79 19.29 20.16 16.50 22.98 15.00 28.43 29.18 30.89 23.74 21.95 24.37 26.37 27.10 26.18 22.61 28.18 20.36 34.84 34.73 35.82 28.41 29.10 29.82 31.72 33.34 33.33 27.43 33.93 27.06 38.22 38.23 32.39 33.63 34.17 35.95 38.10 39.30 32.92 37.52 33.64 37.63 38.00 38.60 40.25 40.81 43.07 38.19 41.88 37.91 44.91 42.04 N 1574 925 561 496 385 182 1017 777 883 514 362 915 EUBOIKOS GULF 23.12 22.66 14.37 13.00 17.94 18.77 22.06 22.1 1 22.62 26.90 19.30 19.00 28.69 28.32 23.68 23.04 23.87 28.24 27.83 27.84 28.15 32.95 26.18 26.47 34.58 35.16 28.97 28.77 29.25 34.03 32.30 31.87 33.47 37.07 32.50 32.24 39.70 40.42 35.10 34.81 34.85 38.98 37.37 37.19 39.07 41.16 37.65 38.03 44.36 45.65 39.83 39.93 39.46 43.83 41.35 41.85 43.68 45.64 42.34 43.20 49.45 44.07 44.32 43.83 44.54 46.26 47.36 51.60 47.77 47.51 48.40 48.42 47.14 48.08 51.58 50.45 53.70 43.58 N 1841 2632 1630 1357 867 1099 485 435 452 464 749 570 Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov 94 94 94 95 95 95 95 95 95 95 95 95 95 95 30.19 33.00 27.00 32.84 28.00 26.33 24.06 25.77 27.31 33.71 36.75 33.50 37.19 34.55 33.20 28.50 31.16 30.1 1 38.51 38.26 43.67 38.59 37.1 1 35.49 36.65 33.82 41.71 44.70 43.67 39.93 43.93 39.31 46.67 401 41 84 107 94 135 362 578 235 29.00 34.00 34.94 24.53 25.86 25.00 28.83 22.00 28.29 28.23 28.49 34.34 34.65 35.00 39.85 44.70 32.24 32.67 37.66 34.24 29.43 34.33 33.89 34.46 39.60 38.55 39.49 37.76 38.12 41.55 38.36 35.81 39.84 39.30 39.96 44.67 44.34 41.99 43.02 46.51 42.63 40.08 44.07 44.20 44.26 45.84 47.49 46.80 44.35 49.00 48.00 48.00 50.82 48.59 51.61 106 99 153 210 283 142 301 172 159 619 339 251 390 21.35 23.00 25.39 17.38 15.33 16.00 24.47 23.68 21.80 23.78 23.38 26.67 21.84 22.01 26.83 27.79 29.49 25.04 19.69 22.75 31.22 32.07 27.90 28.24 27.66 33.40 27.54 27.42 30.73 32.50 32.72 29.55 25.81 26.57 35.60 37.1 1 33.17 33.28 32.20 32.50 32.98 32.89 29.47 30.49 40.00 37.72 38.15 36.63 36.41 37.67 36.68 32.49 34.74 42.13 42.78 40.73 46.14 504 409 478 430 309 490 393 373 587 949 486 310 615 289 14.42 22.64 15.59 18.50 18.00 18.50 19.00 21.63 15.00 21.1 1 30.28 23.47 25.02 24.85 24.54 24.21 23.09 24.29 26.1 1 19.00 27.37 34.63 30.37 31.25 29.16 29.23 29.00 29.68 30.64 31.57 26.38 32.68 39.54 34.87 35.04 34.60 33.85 33.83 34.32 34.61 35.88 32.14 41.67 42.33 39.12 39.04 39.13 38.20 39.10 39.76 36.17 43.45 42.58 42.00 43.20 40.13 45.00 103 370 345 107 180 332 435 1176 490 534 340 25.75 20.91 15.33 30.74 23.58 14.67 16.00 19.06 17.67 21.42 23.24 23.18 18.00 13.64 30.53 26.30 23.44 36.69 33.19 26.22 23.71 23.87 21.61 27.50 27.37 29.55 24.24 18.33 34.26 31.88 28.57 38.69 32.58 28.39 31.39 28.45 33.26 33.27 34.18 29.92 24.21 40.24 35.76 33.75 43.00 40.12 33.92 36.12 34.45 37.38 38.59 38.67 34.58 28.79 39.90 38.43 37.94 39.46 38.30 41.24 39.08 33.55 42.07 42.92 36.60 692 540 774 619 393 317 480 647 586 740 939 500 530 592 14.44 20.59 16.10 20.88 17.98 18.04 18.58 15.33 21.21 21.43 27.20 22.63 25.94 25.06 24.29 24.23 22.48 27.60 27.85 33.84 29.29 31.71 32.87 32.33 32.00 28.10 34.24 33.93 38.1 1 34.00 37.17 37.77 37.42 37.97 33.52 39.62 38.39 41.66 41.61 41.02 42.05 38.70 451 542 200 916 841 598 708 624 699 27.02 21.24 21.78 17.29 15.77 21.42 22.24 17.52 24.51 24.00 21.64 23.28 37.23 26.05 28.05 21.53 21.91 28.04 29.27 24.89 30.37 31.31 28.54 29.86 42.98 32.47 31.86 27.10 28.45 34.33 34.73 30.48 35.92 35.56 35.17 35.52 47.25 37.59 36.97 32.33 34.12 40.02 38.94 35.61 41.50 41.14 40.00 40.10 53.84 42.96 41.73 37.39 39.94 43.68 43.21 41.24 45.78 45.18 45.52 45.12 46.75 46.59 41.28 43.54 48.02 50.58 45.44 49.90 48.96 50.71 46.76 50.37 53.42 1354 590 416 456 838 448 526 1937 772 434 271 635 T ABLE

4. – Identified mean lengths-at-age of female

N. norvegicus

using Bhattacharya’

s method and number (N) of examined individuals from October 1993 to November 1995 in the dif

ferent studied

(10)

93 93 93 94 94 94 94 94 94 94 94 94 Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep A TLANTIC 28.99 27.27 26.63 25.75 29.58 21.01 20.64 24.32 25.07 28.98 30.47 33.99 31.32 30.33 30.82 33.78 27.86 28.24 32.42 29.60 32.87 35.27 39.04 40.20 33.90 34.60 39.18 32.01 33.85 38.30 34.25 37.28 41.18 45.71 38.59 39.16 46.37 35.92 41.61 45.36 38.13 41.53 45.08 51.04 45.59 45.60 40.21 46.67 51.44 43.78 46.69 49.55 53.07 44.86 N 256 24 83 44 74 153 328 227 985 64 164 ALBORAN 33.88 31.92 29.26 28.00 32.98 24.86 24.44 28.48 28.47 26.92 23.97 28.57 43.1 1 45.42 34.77 33.92 38.05 32.46 32.62 39.29 38.67 40.08 31.17 33.01 48.67 50.57 39.14 37.30 42.93 38.50 39.60 44.31 44.10 46.13 37.26 38.93 54.92 49.08 41.62 49.72 44.33 44.29 51.33 50.76 52.22 42.67 46.10 48.33 56.15 51.37 51.04 57.09 47.25 55.02 52.57 58.75 N 710 186 202 56 289 650 644 923 601 798 205 191 CA T ALAN 22.66 23.15 23.52 22.49 21.02 19.69 19.10 23.23 22.46 22.14 22.16 22.43 26.84 27.1 1 27.55 27.77 25.93 25.01 23.36 29.08 30.57 29.41 26.69 27.13 30.43 30.46 33.72 33.56 31.13 30.83 31.48 32.29 34.45 33.39 32.94 31.69 34.27 35.27 38.05 39.32 36.67 34.92 36.37 43.98 39.12 36.93 38.25 38.46 42.56 45.07 42.49 24.20 49.81 54.49 N 9 1 655 526 302 554 281 1196 630 1100 355 237 356 LIGURIAN 16.41 28.07 34.33 15.08 28.65 29.50 24.02 33.69 39.54 19.60 34.10 36.38 29.35 39.1 1 46.49 23.63 37.76 42.81 33.84 45.01 29.56 44.32 40.82 33.74 48.19 53.56 38.55 44.15 47.90 54.56 N 240 437 389 87 460 173 TYRRHENIAN 19.18 17.69 15.63 20.05 11.44 18.72 13.82 19.26 16.18 20.32 20.09 18.05 24.77 22.96 23.94 25.77 20.47 25.86 19.44 22.43 22.19 29.30 27.01 21.69 31.97 27.73 29.62 30.24 31.91 31.42 26.85 30.88 27.54 34.92 34.88 27.10 38.19 32.76 35.52 33.76 44.56 36.96 32.87 36.15 35.41 45.13 44.99 34.39 44.55 36.54 38.16 38.97 44.73 46.18 45.13 51.55 N 447 712 547 474 761 622 824 241 340 398 439 250 ADRIA TIC 12.25 24.54 15.94 16.38 15.74 17.74 19.91 19.54 18.24 17.37 23.95 13.97 22.54 28.93 25.67 23.08 22.52 22.80 26.93 28.05 22.50 22.73 28.07 20.83 26.64 33.97 30.86 28.19 28.26 27.93 31.12 33.30 30.68 29.37 32.04 26.43 35.90 45.45 36.47 32.66 32.65 32.86 35.04 37.26 34.73 33.64 36.03 32.87 45.43 49.56 37.91 38.1 1 38.50 39.79 48.57 42.80 39.64 40.36 37.18 41.82 42.52 44.96 46.22 42.77 48.95 N 1574 925 561 496 385 182 1017 777 883 514 362 915 EUBOIKOS GULF 12.42 26.87 13.90 13.86 17.82 18.06 22.84 18.68 21.31 15.39 20.87 19.18 24.30 37.30 24.88 24.67 23.85 25.51 30.62 22.37 26.84 25.80 26.65 25.93 29.71 46.53 30.99 30.47 29.63 32.43 35.86 27.88 32.98 30.10 34.71 31.97 39.20 53.00 37.15 39.32 35.72 38.40 41.61 34.94 39.10 37.42 40.89 39.40 47.06 67.56 44.78 44.81 42.86 43.02 45.64 40.83 44.68 45.42 47.42 45.16 54.12 53.99 52.02 49.23 49.10 50.35 50.00 50.27 53.10 54.71 55.60 58.56 57.65 54.81 61.59 N 1841 2632 1630 1357 867 1099 485 435 452 464 749 570 94 94 94 95 95 95 95 95 95 95 95 95 95 95 Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov 29.48 26.08 26.87 25.58 20.86 27.80 22.47 27.28 27.10 32.83 32.76 33.10 33.06 28.34 32.87 26.50 31.96 30.82 37.37 36.76 28.49 37.55 34.73 36.64 33.69 36.99 34.78 41.41 41.07 44.61 42.30 39.34 40.59 37.63 43.51 39.09 48.52 46.52 53.61 45.72 44.51 44.72 51.05 51.56 54.56 401 41 84 107 94 135 362 578 235 19.27 19.82 34.84 25.37 21.37 24.90 28.62 23.64 28.31 30.09 29.19 34.40 19.48 29.24 34.12 44.50 37.54 25.70 38.05 39.02 28.51 34.14 34.87 34.69 43.58 36.13 37.83 41.69 46.62 36.83 43.12 45.42 35.48 40.16 41.94 41.68 47.82 46.73 44.52 49.38 42.55 50.27 51.05 42.01 43.87 46.61 45.92 53.70 52.50 47.87 47.92 51.89 49.63 55.06 51.66 106 99 153 210 283 142 301 172 159 619 339 251 390 21.55 19.13 20.69 17.50 18.58 18.80 21.98 23.59 21.62 23.47 18.17 22.95 28.24 29.76 26.09 22.94 24.74 24.18 25.04 22.28 26.04 31.38 27.31 27.53 23.93 26.99 27.19 25.27 30.53 25.94 29.27 29.1 1 29.55 25.54 30.91 35.63 31.97 33.84 28.00 32.16 34.28 28.75 39.59 29.15 38.75 38.22 35.00 29.98 35.14 40.80 36.71 38.1 1 33.87 40.61 43.68 33.28 32.76 34.46 42.09 45.56 41.05 42.44 41.58 39.31 41.74 44.94 504 409 478 430 309 490 393 373 587 949 486 310 615 289 13.53 12.39 16.10 12.44 18.25 17.23 22.77 21.94 21.78 12.38 14.26 21.35 21.98 23.22 19.17 24.94 25.40 27.85 28.01 25.63 24.25 19.45 28.47 31.56 31.00 24.43 28.82 31.99 36.39 32.83 29.99 30.1 1 25.89 33.68 39.81 42.96 31.96 34.63 39.01 43.82 28.24 33.86 35.46 31.40 41.75 36.52 40.46 54.56 43.31 41.44 41.37 37.42 42.80 44.32 56.23 103 370 345 107 180 332 435 1176 490 534 340 25.04 14.08 12.73 14.42 14.44 14.46 15.84 18.46 20.54 21.22 11.44 15.29 18.87 15.03 30.21 26.30 22.68 30.95 24.98 25.17 22.85 24.14 29.47 29.90 26.36 22.93 24.37 23.28 35.56 33.42 27.90 43.51 33.39 32.67 29.06 32.19 35.79 35.25 33.15 29.89 29.30 27.85 41.90 40.79 33.32 39.99 41.41 34.77 37.45 44.64 42.20 41.99 34.06 34.15 33.19 39.66 46.66 40.60 38.80 39.13 43.46 51.56 45.53 692 540 774 619 393 317 480 647 586 740 939 500 530 592 14.42 17.36 16.95 21.02 19.69 18.05 19.25 14.98 19.65 21.01 25.52 22.64 26.91 24.06 23.62 23.24 21.65 24.12 27.32 29.98 27.76 31.09 32.19 32.14 28.02 26.28 27.54 33.83 34.61 30.49 36.1 1 38.48 38.30 32.16 31.85 33.61 39.35 40.1 1 35.43 42.25 50.56 44.14 37.01 36.57 39.39 40.59 41.40 41.76 48.73 451 542 200 916 841 598 708 624 699 26.88 15.29 21.04 16.94 15.64 20.07 21.06 18.90 20.13 26.37 21.99 18.42 35.40 20.79 26.98 21.76 25.22 26.09 26.50 23.56 24.09 34.02 27.81 23.21 40.09 25.08 30.76 26.54 32.89 33.30 33.28 29.10 29.29 40.23 33.87 28.18 45.58 30.50 35.51 32.13 40.50 38.48 37.82 34.85 34.99 46.25 38.88 34.93 52.94 37.33 40.25 39.69 45.14 43.81 42.61 40.36 40.02 50.37 43.84 40.92 42.10 46.14 47.82 50.96 49.69 46.90 44.93 45.1 1 54.55 48.88 47.72 47.57 51.62 51.07 51.50 56.56 52.74 1354 590 416 456 838 448 526 1937 772 434 271 635 T ABLE

5. – Identified mean lengths-at-age of female

N. norvegicus

using MIX method and number of examined individuals from October 1993 to November 1995 in the dif

(11)

between areas, the examination of the closeness of the mean lengths-at-age in the various months showed that for males L, T, A, G and P were close, whereas C was more distant and M even more so. For females, A, G and L constituted a cluster, and T, P and C another, while M remained the only rep-resentative of a separate cluster. Figure 5 presents the results of this analysis. The comparison of the mean lengths-at-age, expressed by the MAPD val-ues, showed similar results. For both sexes, most of the larger MAPD values were between, on the one hand, M and C, and on the other, A, G, P, L and T. The largest MAPD values of all was between M and C. For males, all distances between A, G, P, L and T were low. For females, however, the MAPD distances distinguished two groups, one comprising A, G and L, and the other including T, P and C. T and P were closer to the first group than C. The same was also found for M. The results of the MAPD values are presented in Table 6.

The estimation of the growth parameters in both approaches proved to be difficult. The estimates were not always acceptable. In some cases, the

val-ues of L∞were lower than expected (e.g. in Table 7

the case of Malaga) considering the maximum sizes present in the catches (Table 1). On the other hand, the application of the von Bertalanffy growth model presented difficulties. In many samples, the

FIG. 4. – Mean lengths-at-age estimated by MIX and Bhattaharya’s method BHAT (only males of Catalan Sea are presented here as an example).

LTGAPCM

GAL MPTC

FEMALES

MALES

25

20

15

10

5

0 A

verage Linkage

FIG. 5. – Cluster analysis using the Average Linkage method for

the examination of the distances between the mean lengths-at-age of the 4 year old age group of the various areas for each sex sepa-rately. (P), Atlantic; (M), Alboran Sea; (C), Catalan Sea; (L), Ligurian; (T), Tyrrhenian; (A), Adriaric and (G): Euboikos Gulf.

(12)

TABLE7. – Estimated growth parameters of N. norvegicus for each area and sex, using the Gauss-Newton method and SAS program. The table includes also the number of pairs of age-length values (n) used to produce the estimates and the Mean Square Error

(MSE) value for the fitting.

Males AREA L∞ k to n MSE Atlantic (P) 71.3 0.10 -2.45 59 1.12 Alboran Sea (M) 78.4 0.17 -0.38 39 0.85 Catalan Sea (C) 72.9 0.14 -1.43 40 1.96 Tyrrhenian Sea (T) 80.8 0.13 0.07 61 1.63 Adriatic Sea (A) 71.4 0.11 -1.18 88 3.29 Ligurian Sea (L) 65.2 0.16 -0.96 32 2.94 Euboikos Gulf (G) 82.4 0.12 -0.01 79 2.55 Females AREA L∞ k to n MSE Atlantic (P) 62.4 0.14 -1.19 39 1.79 Alboran Sea (M) 59.4 0.20 -0.92 49 4.99 Catalan Sea (C) 54.9 0.18 -1.36 38 2.06 Tyrrhenian Sea (T) 69.4 0.12 -0.64 46 2.29 Adriatic Sea (A) 68.0 0.14 -0.21 30 1.99 Ligurian Sea (L) 54.5 0.22 0.03 37 2.78 Euboikos Gulf (G) 75.8 0.12 -0.11 79 2.76 TABLE6. – Estimated values for the absolute difference of the

mean lengths-at-age expressed as a percentage of their mean (MAPD) between the different studied areas. (P), Atlantic; (M), Alboran Sea; (C), Catalan Sea; (L), Ligurian; (T), Tyrrhenian; (A), Adriaric and (G): Euboikos Gulf. *for the symbol of the area see

in the text. Males M* C T P A G C 25.0 T 10.8 15.0 P 16.6 9.1 6.2 A 13.2 12.9 6.2 7.0 G 12.5 13.2 5.3 6.6 5.1 L 9.6 16.5 5.5 9.3 8.0 6.8 Females M C T P A G C 21.7 T 16.4 6.3 P 16.8 6.2 4.6 A 7.9 13.8 9.9 8.9 G 8.9 13.6 10.0 9.7 6.4 L 10.6 11.7 7.0 7.0 6.4 5.5

TABLE8. – Growth parameters of N. norvegicus estimated by the FISHPARM program, obtained from the analyses of one or more

selec-ted months as well as the analysis of all months for each area and sex. The mean square error (MSE), the growth performance index φ´ and

the number of analyzed months are also presented.

Males Females

AREA L∞ k to MSE φ Months L∞ k to MSE φ Months

Atlantic (P) 78.9 0.14 -0.56 1.30 2.94 3 71.3 0.12 -1.15 0.61 2.79 3 83.4 0.13 -0.33 0.24 2.96 1 70.7 0.12 -1.36 0.29 2.78 1 158.3 0.04 -2.25 4.03 3.00 20 90.4 0.07 -2.11 2.29 2.76 20 Alboran Sea (M) 85.2 0.14 -0.80 0.84 3.01 6 75.5 0.14 -0.87 1.13 2.91 6 86.8 0.14 -0.84 0.13 3.02 1 72.6 0.16 -0.87 0.19 2.93 1 91.3 0.12 -1.08 1.29 3.00 25 93.9 0.09 -1.61 1.94 2.90 25 Catalan Sea (C) 85.7 0.10 -0.52 2.99 2.88 7 71.4 0.12 -0.53 1.05 2.79 3 86.8 0.10 -0.30 0.09 2.87 1 67.0 0.15 -0.33 0.03 2.73 1 94.2 0.09 -0.81 3.06 2.89 26 171.1 0.03 -1.80 1.67 2.99 26 Ligurian Sea (L) 86.5 0.12 -1.05 5.31 2.94 6 67.6 0.14 -0.97 1.95 2.80 3 83.2 0.12 -0.87 0.04 2.92 1 63.2 0.15 -0.89 0.02 2.79 1 89.0 0.11 -1.08 4.88 2.94 17 77.4 0.11 -1.32 1.48 2.81 17 Tyrrhenian Sea (T) 79.9 0.13 -1.02 1.30 2.91 9 70.0 0.12 -1.08 1.84 2.78 6 81.6 0.13 -0.89 0.08 2.94 1 65.0 0.15 -0.76 0.28 2.80 1 99.8 0.09 -1.39 1.72 2.94 26 87.8 0.08 -1.26 2.75 2.82 26 Adriatic Sea (A) 83.3 0.11 -1.24 1.60 2.90 6 68.5 0.14 -1.02 1.37 2.83 7 81.5 0.11 -0.95 0.03 2.87 1 67.0 0.14 -0.88 0.42 2.80 1 120.8 0.06 -1.92 4.56 2.95 21 81.8 0.10 -1.36 3.28 2.84 21 Euboikos Gulf (G) 86.4 0.11 0.89 2.97 2.93 6 75.7 0.13 -0.89 1.94 2.87 9 82.7 0.12 -0.95 0.80 2.91 1 73.9 0.14 -0.47 0.13 2.89 1 93.2 0.10 -1.10 2.79 2.93 24 90.3 0.09 -1.27 2.27 2.88 24

(13)

number of age groups identified was small and included only young ages. In other cases the incre-ments between adjacent modes did not decrease from smaller to larger sizes, showing no decelera-tion in growth, a condidecelera-tion necessary for the appli-cation of the model.

The growth parameter estimates for each area and sex using the Gauss-Newton method are pre-sented in Table 7. In Figure 6 are shown the plots of

the simultaneous confidence limits of L∞and k. In

males, confidence limits were quite narrow, but always overlapped between areas. This was also apparent for females, for which confidence limits were very wide. Table 8 presents the growth para-meters values obtained using the FISHPARM pro-gram as well as the growth performance index φ estimates. The results of three different analyses by the latter method are included in Table 8; those from a selected (as mentioned above) number of months, those from one selected month and finally

those using all months. The growth parameters derived by the first two analyses (analyses based on one or more selected months) were quite close, probably because they were estimated on more or

less the same principle. On the other hand, the L∞

estimates in the third analysis were always greater (and k values always lower) than those of the first two analyses. This was expected since in the first two analyses only months with decreasing growth increments were selected. The results of the third one, even if they had the advantage of being esti-mated from more data, were not taken into account, since for the females of all studied areas (except for

L) L∞ was much greater than 80 mm and for two

areas in males (P and A) much greater than 100 mm. Moreover, the second analysis presented the disadvantage of using a lot a conventions and con-sequently of using very few data. For this reason, it was not taken into account but only presented for comparison purposes, as the previous analysis.

100.0 90.0 80.0 70.0 60.0 50.0 40.0 30.0 20.0 100.0 90.0 80.0 70.0 60.0 50.0 40.0 30.0 20.0 x x x x x + + + + + x x x x x x x x x + + + + + x x x x x x x 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 FEMALES MALES x x + P M T A L G

L

∞∞

K

C

FIG. 6. – Plots of the estimated growth parameters L∞and k and 95% confidence limits are

expressed in the form of an ellipse. The height of the ellipse corresponds to the confidence limits for k and the width to the confidence limits for L∞. P: Atlantic, M: Alboran Sea, C:

(14)

The pairwise comparisons of growth parame-ters, obtained by the Gauss-Newton method, between the different areas (within the same sex)

using Hotelling’s T2_{test were all significant at the}

0.05 level. The visual examination of the growth curves obtained by the FISHPARM program (analysis of selected months), presented in Figure 7, gives an indication of growth differences between the areas. It was observed that for males the curves of M and B were very different from each other and quite different from those of all other areas, which were close. With respect to

females, M showed the most different curve, fol-lowed by G. The curves of the other areas seemed to be close. More information regarding the growth curves of the different areas (derived by the FISHPARM program) could be obtained from the φ values presented in Table 8. For males, it was observed that the highest value belonged to M (3.01) and the lowest to C (2.88). Other areas showed values ranging from 2.90 to 2.94. For females, the highest value was again in M (2.91), followed by G (2.87). The other areas presented values ranging from 2.78 to 2.83.

(15)

DISCUSSION

Length frequency analysis is influenced by many factors. In the present work, a great effort was made to apply this analysis in the most correct way. Since the length-based analysis is applied to length fre-quencies, their structure is of great importance. Some problems that may affect the structure of length frequencies are as follows. (A) The size of the sample (see for example MacDonald and Pitcher, 1979; Pauly, 1984 cited in Hoening et al., 1987; Erzini, 1990); for this purpose, large samples were collected every month in most cases (>200 individuals per sex each month). (B) Poor sampling of small and large length classes, affecting their rep-resentation in the frequencies; this explains the dif-ficulty of identifying the youngest and oldest age groups in this work. From Table 1, the minimum and maximum sizes caught in each area could be compared. (C) Gear selectivity and fishing mortali-ty conditioned by the mesh size are factors influ-encing the structure of a frequency distribution and consequently the length-based analysis and growth estimation. It must be noted that the mesh size for the sampling in the present work varied between areas. More specifically, in the samples from P (mesh size 55mm), it was difficult to identify the age groups 0+ and 1, while in the samples from G (mesh size 32mm), the youngest age groups were better represented and consequently easier to distin-guish. (D) The purity of the samples; it is known that the length structure of Nephrops populations change between adjacent areas (Bailey and Chapman, 1983; Anon., 1988; Tuck et al., 1997). For the present work, the samples were collected from a single station in most cases with the excep-tion of T and G, where the sampling covered more than one station, although in a restricted area. (E) The size of class interval (Erzini, 1990); in the pre-sent work a 1mm interval was used by both teams, as has already been done by previous authors (e.g. Tully et al., 1989). The same class interval has also been approved by Castro (1990; present work) with simulated data and by Mytilineou and Sardà (1995) with original data as the most adequate for the N.

norvegicus length frequencies.

Another problem in length frequency analysis is the uncertainty if the different identified compo-nents correspond to the real number of age groups composing the populations under study. In the pre-sent work, the consideration of the age groups by both teams was done using a series of criteria (see

methods) in order to avoid or limit the subjective interpretation of the results. A confirmation of the accuracy of our results was the continuing presence of some components in the length distributions dur-ing the two years of sampldur-ing. This fact gave the possibility to follow the different year classes (especially three of them) over time by means of the modal progression. Moreover, the consideration of the different components as age 0+, 1, 2 etc. could be considered reasonable if compared to the infor-mation concerning the larval cycle of the species (Farmer, 1973) and the age groups estimated by other authors (e.g. Farmer, 1973; Hillis, 1979; Tully

et al., 1989).

Generally, the estimation of the von Bertalanffy growth model parameters requires deceleration of the growth with time. This poses a problem for the adequacy of the von Bertalanffy model, since this was not observed in many cases. If this model is not appropriate for this species, as suggested by studies with simulated data (Castro, 1992; Castro et al., 1998), the removal of data that does not fit the von Bertalanffy model would lead to erroneous results. An alternative could be the use of another growth model, and in fact during some stages of this work the Gompertz model (RZF) was also used. However, the authors suggested that the radical approach of rejecting the von Bertalanffy growth model was not appropriate. First, it is biologically very difficult to consider non-decelerating growth in adult phases. The indications that this may happen in some periods of the life of slow growing decapods are not sufficient yet for retiring the von Bertalanffy growth curve. Such a situation would need evidence from natural populations that is not available. Second, the von Bertalanffy growth curve, even if it is not the best possible model, is easy to apply in fisheries models, it has parameters with biological meaning and easy interpretation, and it is without doubt the most widely used growth model in fisheries. The estimated parameters, even with questionable absolute values when resulting solely from length frequency analysis, be useful for comparative purposes. Since the objective of this project was a comparative study among several areas of the Mediterranean and adjacent Atlantic, the estimation of the von Bertalanffy growth para-meters was kept as a means for the data analysis.

In the present work, length frequency analysis presented some difficulties. In certain cases, sam-ples were small and/or their structure did not always permit the detection of the young or old age

(16)

groups. Moreover, in any area the whole number of age groups present could not be identified in the length frequency distributions. The progression of modes showed some problems too; the modes pre-sented a variability over time, probably caused by the individual moulting variability. The estimation of the growth parameters was also proved difficult. Their direct estimation, based on the original data using ELEFAN I program and Shepherd’s (1987) method, was found unreliable. In addition, modal progression analysis was not considered so ade-quate for growth estimation (Castro et al., 1998). The indirect estimation of the growth curve from the mean lengths-at-age, using the Gauss-Newton method and the FISHPARM program provided bet-ter results, although they were not always

accept-able. The estimation of L∞became difficult, and if

L∞is underestimated, k will necessarily be

overesti-mated with all the consequences in subsequent use in assessment models. The alternative to this

prob-lem would be to force L∞ to be a chosen value. In

this case, the estimation of k would have little meaning, as discussed by Knight (1968). The prob-lem was probably related to the absence of infor-mation for older ages. If only younger ages are pre-sent, large standard errors result for the parameters. This is expressed in Figure 6, where males have estimates of growth parameters with much narrow-er confidence limits than females. This may be the result of better representation of older ages in the catches of males, since no significant behavioural changes during the year, absence of synchronised moulting, slow progression of mean lengths-at-age and faster growth characterise this sex. All these aspects make the estimation of male growth para-meters easier and more accurate. Females show very wide ranges for the estimated parameters, a sign of poor results for this sex. Some of the facts that contribute to this may include behavioural aspects and difficulty in separating ages during the moulting season. Behavioural patterns affect the population structure since the catchability to the gear decreases for ovigerous females hidden in their burrows. For age classes that do not reach 100% maturity, this could result in bias in the esti-mation of mean lengths-at-age during the ovigerous season. After hatching of the larvae, mature females are well represented in the catch, but this is also the period when moulting occurs. During the moulting season it will be very difficult to separate ages because a given individual starts in one mode and ends in the next one, while still belonging to

the same age group. When attributing ages to the identified groups along the moulting period, bias will necessarily occur. We believe that these diffi-culties are shown in the wide confidence limits

both for L∞ and k of females (Fig. 6). The

differ-ences in the biology of the two sexes is a fact already discussed by various authors (e.g. Farmer, 1973; Charuau, 1975; Hillis, 1979; Sardà, 1985; Anon. 1988). Because of all the above mentioned problems, the approach for the estimation of growth parameters should be the identification of modes within each sample and the combination of data that are not greatly affected by factors such as moulting and behaviour. If growth parameters are used for stock assessment and management purpos-es, perhaps ranges of valupurpos-es, not point estimatpurpos-es, should be taken into consideration.

With respect to the differences between areas, from all the analyses (distance between the mean CL of the monthly samples, distance between the mean lengths-at-age and MAPD) it was obvious that, the majority of the areas gave very close results for males. Only the Alboran and the Catalan Seas dif-fered from the other areas with the greatest distance between them. The results for females were not so clear. The examination of the similarity of the differ-ent areas based on the examination of the mean CL, showed that the Alboran Sea was very different from all other areas (Fig. 1). When the analysis was based on the mean length-at-age, Alboran Sea constituted again a separate group, but two more groups were also detected; one consisting of Adriatic, Ligurian and Euboikos Gulf and the other of Catalan, Tyrrhenian and Atlantic (Fig. 5). This was also found by the examination of MAPD values (Table 6).

The differences in growth parameters between areas were found to be statistically significant in all cases for both sexes. However, it was doubtful if they were also biologically significant. For male N.

norvegicus (excluding the Ligurian Sea which had

quite less information available), all estimates of L∞

obtained by the two approaches (Gauss-Newton and FISHPARM) were between 71 and 87 mm (Table 7 and 8), a narrow range taking in consideration the differences in population structure shown in the studied areas and the results of other researchers. Bailey and Chapman (1983) found for two Nephrops populations a wider range for males’ L (68.9 mm and 46.6 mm). Even for the sub-populations of a small area, Tuck et al. (1997) found a greater difference (from 45.3 mm to 65.1 mm) comparing with that of the present study. The plotting of the growth curves

(17)

revealed that for males all curves were very close, except in the case of Alboran and Catalan Seas (Fig. 7). This was also observed by the φ values (Table 8) with the highest value corresponding to the Alboran and the lowest to the Catalan Sea. In the case of females, for which the results were not so accurate, the curve for the Alboran sea was the most distant, followed by Euboikos Gulf (Fig. 7). They generally were close to each other, however, Alboran Sea showed the greatest value, followed by Euboikos Gulf. For the other areas the φ values were found to be very similar. Furthermore, it must be pointed out that the results from the visual comparison of growth curves coincided with those from the examination of the mean CL among areas. This is in accordance with the results of Tuck et al. (1997), which found

that the mean CL is positively correlated to L∞and

therefore it could be used as an indicator of growth variability.

In summary, the results of this work generally lead us to suggest that, differences exist in the growth pattern of N. norvegicus between the differ-ent studied areas of the Mediterranean Sea and the adjacent Atlantic, although these could not be con-sidered very important, with the exception of one area. It was clear that N. norvegicus from the Alboran Sea presented a great difference in its growth pattern compared with all other areas. For the Adriatic, Tyrrhenian, Ligurian and Atlantic great similarities were found in all analyses. Very close to them was also the Euboikos Gulf (with respect to the males) and the Catalan Sea (with respect to the females). Differences or similarities between areas are probably related to environmental factors (sedi-ment, temperature, etc.) and biological factors (den-sity, availability of food, etc.), the subject of future work by the authors.

ACKNOWLEDGEMENTS

The authors wish to express their gratitude to the EC, DG. XI, (MED92/008), as well as Dr. Sardà, Dr. Froglia, Dr. Relini and Dr. Biagi for the offer of the length frequencies data. In addition, we want to thank the two anonymous reviewers as well as Dr. I. Tuck and Dr. O. Tully for the critical review of the manuscript and their useful remarks as well as Prof. C. Richardson for statistical advice and Dr. C.-Y. Politou for useful comments. Finally, we would also like to thank Mr. J. Dokos and Mrs V. Lambropoulou for drawing some of the figures.

REFERENCES

Anonymous. – 1988. Report of the study group on Nephrops.

ICES. CM 1988/K:29.

Bailey, N. and C.J. Chapman. – 1983. A comparison of density, length composition and growth of Nephrops populations off the west coast of Scotland. ICES. CM 1983/K: 42.

Bernard, D.R. – 1981. Multivariate analysis as a mean of compar-ing growth in fish. Can. J. Fish. Aquat. Sci., 38: 233-236. Bertalanffy, L. von. – 1957. Quantitative laws in the metabolism

and growth. Q. Rev. Biol., 32: 217-231.

Bhattacharya, C.G. – 1967. A simple method of resolution of a distribution into Gaussian components. Biometrics, 23: 115-135.

Castro, M. – 1990. The use of length frequency analysis for esti-mation of the age structure of the catch of Nephrops

norvegi-cus. Rapp. P-v Reun. Int. Explor. Mer.

Castro, M. – 1992. A methodology for obtaining information on the age structure and growth rates of the Norway lobster,

Nephrops norvegicus (L.) (Decapoda, Nephropoidea). Crustaceana, 63: 29-43.

Castro, M. – 1995. The use of length-frequency analysis for esti-mating the age structure of the catch of Nephrops norvegicus (Crustacea: Nephropidae). ICES mar. Sci. Symp., 199: 301-309.

Castro, M., Ch. Mytilineou. and P. Gancho. 1998. – Methodological considerations concerning the use of length frequency analysis for growth studies in the Norway lobster,

Nephrops norvegicus (L.). Sci. Mar., 62 (Supl. 1): 61-69.

Charuau, A. – 1975. Croissance de la langoustine sur les fonds du Sud-Bretagne. ICES. Shell. Comm., CM. 1975/K: 11, 15pp. Conan, G. – 1978. Average growth curves and life history in a

Nephrops norvegicus population from Northern Bay of

Biscay. ICES. CM 1978/K: 18, 56pp.

Farmer, A.S.D. – 1973. Age and growth in Nephrops norvegicus (Decapoda: Nephropidae). Mar. Biol., 23: 315-325.

Figueiredo, J.M.. – 1984. Attempts to estimate growth and instan-taneous matural mortality of Nephrops norvegicus off the Portuguese coast. ICES. CM 1984/K:28, 19 pp.

Erzini, K. – 1990. Sample size and grouping of data for length-fre-quency analysis. Fish. Res., 9: 355-366.

Gayanilo, F.C., J.M. Soriano and D. Pauly. – 1988. A draft guide to the COMPLEAT ELEFANT. ICLARM Software Project 2, 65 pp.

Gayanilo, F.C., P. Sparre and D. Pauly. – 1996. The FAO-ICLARM Stock Assessment Tools (FiSAT) User’s Guide. FAO

Computerized Information Series, Fisheries. Rome, 126 pp.

Hanumara, R.C. and N.A. Hoenig - 1987. An empirical compari-son of a fit of linear and non-linear models for seacompari-sonal growth in fish. Fish. Res., 5: 359-381.

Hillis, J.P.- 1979. Growth studies on the prawn, Nephrops

norvegi-cus, Rapp. P-v. Reun. Int. Expor. Mer, 1975:170-175.

Hoening, J.M., J. Csirke, M.J. Sandres, A. Abella, M.G. Andreoli, D. Levi, S. Ragonese, M. Al-Shoushani and M.M. Ae-Musa. – 1987. Data aquisition for length-based stock assessment: Report of working group I. In: D. Pauly and G.R. Morgan (Ed.), Length-based methods in Fisheries Research. ICLARM Conf. Proc. 13, February 1985, Mazzara del Vallo, Sicily, Italy. International Centre for Living and Resources Management, Manila Philippines, and Kuwait Institute for Scientific Research, Safat, Kuwait, pp. 343-352.

Knight, W. – 1968. Asymptotic growth: an example of nonsense disguised as mathematics. J. Fish. Res. Board Canada 25: 1303-1307.

MacDonald, P.M.D. and P.E.J. Green. – 1988. User’s Guide to

pro-gram MIX: An interactive propro-gram for fitting mixtures of dis-tributions. Ichthus Data Systems, Hamilton, Ontario, Canada,

29 pp.

MacDonald, P.D.M. and T.J. Pitcher. – 1979. Age groups from size-frequency data: a versatile and efficient method of analysing distribution mixtures. J. Fish. Res. Board Can., 36: 987-1001.

Mytilineou, Ch., A. Fourtouni and C. Papaconstantinou. – 1993. Data on the biology of Norway lobster, Nephrops norvegicus (L., 1758) in the North Aegean Sea. Proceedings of the 4th

Panhellenic Symposium on Oceanography and Fisheries,