PRELIMINARY STUDY ON AGE , GROWTH AND REPRODUCTION OF MUSTELUS MUSTELUS ( ELASMOBRANCHII : CARCHARHINIFORMES : TRIAKIDAE ) INHABITING THE GULF OF ISKENDERUN , NORTH-EASTERN MEDITERRANEAN SEA

Background. Shark populations within the eastern Mediterranean are declining due to overfishing. Unfortunately, our knowledge on the biology of sharks and specifically dogfishes within eastern Mediterranean is extremely limited due to the lack of landings. Mustelus mustelus (Linnaeus, 1758) is one of such deficiently known cartilaginous fishes and the relevant information on its age and growth in the eastern Mediterranean is missing. The understanding of growth and reproductive behaviour is important for management and conservation of this species. The presently reported study was intended to determine age, growth, and some reproductive parameters of Mustelus mustelus collected in the Gulf of Iskenderun, the north-eastern Mediterranean Sea. Materials and methods. In this study, a total of 155 Mustelus mustelus were caught in the north-eastern Mediterranean Sea between March 2012 and October 2015. Fish age was determined based on band counts of sectioned vertebrae. The periodicity of band pair formation was examined using the marginal increment analysis. Growth parameters for all specimens were then determined by fitting the observed and the length-at-age data using the von Bertalanffy Growth Equation. The size and age at which 50% of both sexes reached maturity was estimated by a logistic model for maximum likelihood process and the gonadosomatic index was determined. Results. The total length and weight of the specimens examined ranged from 44.3 to 162.6 cm and from 241.3 to 12 060 g, respectively. The presently reported study provides new maximum lengths (162.6 cm) for M. mustelus from the north-eastern Mediterranean. The fish studied represented age groups from 0 through 25. The observed sex ratio was 1 ÷ 1.04 (females ÷ males). Von Bertalanffy growth parameters were estimated to be L∞ = 195.13 cm, W∞ = 20 060 g, K = 0.06, and t0 = −4.27 for all specimens. Length–weight relation was found to be W = 0.0027L 3.0054


INTRODUCTION
Smooth-hound sharks belong to the genus Mustelus (family Triakidae), which includes 34 valid species occurring in all major oceans.Five of those species can be found in the Mediterranean.Mustelus mustelus (Linnaeus, 1758) is a demersal species inhabiting sandy or muddy bottom down to the depths of 150 m (Golani et al. 2006).This species is caught in small numbers by bottom trawl and they have minor commercial value in Turkey (Başusta and Erdem 2000).The most important prey-groups for the species in the north-eastern Mediterranean are crustaceans and fishes (Özcan and Başusta 2016).
Sharks and specifically dogfishes have become an important commodity as a fishery resource in today's world, where alternative seafood supplies have gained importance.In spite of this, there is almost no detailed biological information about this species.It is known that dogfishes have low growth rate, reach sexual maturity at later age, and their fecundity is rather low (Cailliet et al. 1993).Mustelus mustelus is a widespread species, although not abundant.The species may be threatened with extinction in the future (Serena et al. 2009).
Length-weight relations (LWRs) are of high importance to compare life histories of fishes between different areas (Hossain et al. 2013).The results will be useful for sustainable management and protection of the limited stocks in the north-eastern Mediterranean Sea.
The aim of this study was to determine, for the first time, age, growth, length-weight relations, and reproductive biology of M. mustelus in the north-eastern Mediterranean Sea.

MATERIALS AND METHODS
The samples were obtained monthly by gill nets and long-lines between March 2012 and October 2015 in the north-eastern Mediterranean Sea including the Gulf of Iskenderun, Turkey (Fig. 1).Following transportation of the samples to the laboratory, the identification of species was made according to Golani et al. (2006).For each fish total length (TL) and weight were determined to the nearest 0.1 cm or 0.1 g, respectively.Presence or absence of claspers aided sex identification.
A total of 12 vertebral centra were dissected from above the abdominal cavity of 155 Mustelus mustelus, stored frozen (for methodology see Türkmen et al. 2005).Vertebrae were then separated from each other and immersed in warm distilled water (Martin and Cailliet 1988, Cailliet and Goldman 2004, Cailliet et al. 2006, Duman and Başusta 2013).Large vertebrae, exceeding 5 mm in diameter were divided using a gem saw (Ray Tech) with two diamond blades separated by a 0.6 mm spacer (Başusta and Sulikowski 2012).Smaller vertebrae were sanded with a Dremel tool to replicate a sagittal cut (Sulikowski et al. 2005, Başusta et al. 2008).Reading of the vertebral sections was done using a light-reflecting microscope.One growth band was defined as an opaque and translucent band pair that traversed the intermedialia and clearly extended into the corpus calcareum (Fig. 2).
The index of the mean percentage error * (IMPE) was calculated to assess the precision of the age determinations between 2 independent readers.The equation is expressed as follows: where n is the number of fish aged, N = number of times each fish was aged, X ij is the ith age determination of the jth fish, X j is the mean age calculated for the jth fish (Beamish and Fournier 1981).Age determination bias between readers was assessed using an age-bias plot (Campana 2001, Cailliet andGoldman 2004).
The periodicity of band pair formation was examined using the marginal increment ratio (MIR) (Sulikowski et al. 2003).The MIR was calculated as the ratio of the margin between the last and penultimate opaque bands as measured with an optical micrometre.The marginal increment ratio (MIR) was calculated by the following equation of Natanson et al. (1995 where VR is the vertebral radius, and VR n and VR n -1 are the last and penultimate opaque bands, respectively.A sub-sample of 48 vertebrae was selected comprising both juvenile and adult specimens collected nearly every month except August.Kruskal-Wallis one-way analysis of variance on ranks was used to test for differences in marginal increment by month (Cailliet et al. 1990, Sulikowski et al. 2003).
where L t is the expected total length at age t years and L ∞ is the asymptotic mean maximum total length, K is the growth coefficient, and t 0 is the theoretical age at zero length.
The growth in weight was also described by the weight-based von Bertalanffy growth function where W t is total weight at time t, W ∞ is the maximum theoretical weight of fish, and b is power constant of length weight relations.
The relation between length and weight was calculated using the expression where W is the total weight (TW) [g], L is the total length (TL) [cm], a is the intercept of the regression, and b the slope or regression coefficient.The b values of both sexes for Mustelus mustelus were tested by using the Kolmogorov-Smirnov two-sample test, and the b value variation from 3 was tested with the one sample t-test (P < 0.001).The degree of association between the variables was computed by the determination coefficient R 2 (King 1995).
Condition factor K was determined by the formula where W is the total weight, TL is the total length, and b is the coefficient of allometric relation (Bagenal and Tesch 1978).The size and age at which 50% of both sexes reached maturity (TL50% and A50%, respectively) were estimated by a logistic model for maximum likelihood process (Ghorbel et al. 1996, Conrath andMusick 2002).This method utilizes binomial maturity stage (immature = 0, mature = 1), so all specimens that were not capable of breeding (juveniles and maturing specimens) were grouped as immature.
The gonadosomatic index [%] was determined using the formula where W g is the gonad weight [g], and W t is the total weight of fish [g] (Bagenal 1978).

RESULTS
A total of 155 individuals of Mustelus mustelus (76 females, 79 males) were collected during the study.The female ÷ male ratio was found to be 1 ÷ 1.04.Females ranged from 44.3 to 162.6 cm in total length (TL) and 241.3 to 12 060 g in total weight (W).Males ranged from 47.1 to 1490 cm in TL and 315.6 to 8776 g in W. The length differences between females and males were not statistically significant (Student's t-test, P > 0.05).The total length-frequency distribution by sex is given in Fig. 3.The strong linear relation between centrum diameter and TL indicates that the vertebrae provided a continuous record of body growth (R 2 = 0.95) and therefore is a suitable structure for age interpretation in both male and female M. mustelus (Fig. 4).
The age of M. mustelus specimens was determined by vertebral band counts.The age ranged from 0 to 25 years.IMPE, which is the index of mean percentage error determined by two independent age readers, was found to be 3.03% for M. mustelus.This means that the confidence interval for the reliability of estimations was between 5% and 15%, indicating that our ageing method represents a precise approach to the age determination of M. mustelus (Campana 2001, Girgin andBaşusta 2016).
Marginal increment widths were significantly different between all months (Kruskal-Wallis, P < 0.001), with increment growth beginning in January and peaking in July (Fig. 5).This peak was followed by a large decline in marginal increment growth, reaching minimum values in October.As such, these data suggest that a single opaque band is formed annually on the vertebral centra between July and October.Age bias plot between readers for all age estimated is given in Fig. 6.Both readers estimated same age in 122 of 140 samples.The agreement between readers on all age assessment was 87%.The agreement between readers was 89% within 1 year and 95% within 2 years.The difference between readers in the accuracy of age assessment was found insignificant when repeated annuli counts were compared (paired t-test, P > 0.05).
In total, 75 individuals represented this age group (38 females and 37 males) and age 2 was the dominant age group among all (Table 1).Age, total length, and weight distribution of M. mustelus are given in Table 1.Age-frequency distribution by sex is given in Fig. 7.The maximum length at age was determined as 162.6 cm at 25 years of age and 150.2 cm at 17 years of age (Fig. 2).
In this study, the estimated von Bertalanffy Growth Equation (VBGE) was L t = 195.13[1 − e −0.06 (t + 4.27) ] for both sexes combined by considering the length of fish based on the population of Mustelus mustelus in the north-eastern Mediterranean (Fig. 8).The maximum absolute growth of fish in 1-year age group was found to be 8.79 cm; relative growth was calculated as 19.94%.On the other hand, the minimum absolute growth in fish within the 25 years of age was found as 2.08 cm; relative growth was calculated as 1.30%.
The length and weight measurements, sample sizes (n), regression parameters a and b of the LWR, 95% confidence intervals of b, and coefficients of determination (R 2 ) of the M. mustelus are given in Table 2     coefficient of determination (R 2 ) was 0.98 for M. mustelus (Fig. 9).Regression analysis showed that fish length has high significant correlation with weight (R = 0.99, R 2 = 0.98, F 1, 138 = 7372.259,P < 0.001) and it is possible to say that 98% increase in weight was due to length increase.Normality test was done by using the Kolmogorov-Smirnov twosample test.When the t-test results were analysed for the significance of regression coefficients (t-test = 204.916,P < 0.01), it was found that fish-length data could be used with high accuracy to predict fish weight.Condition factor was calculated for all age groups.The highest condition factor was found at the age of 11 (0.3507), while the lowest was found at the age of 16 (0.2517).
Length at 50% maturity was found as 109 cm in females and 92 cm in males for M. mustelus in the northeastern Mediterranean Sea (Fig. 10).Based on length data, the sexual maturity age was found to be 8 years for females and 7 years for males.As a result, it was determined that M. mustelus has reached the ability to reproduce at 7-8 years of age in the north-eastern Mediterranean Sea.
Gonadosomatic index (GSI) values varied from 0.1485 to 2.878 for females, and from 0.1038 to 0.2385 for males.It was observed that mean GSI values were low in September and July, reaching a maximum from January through February (Fig. 11).
In this study, two pregnant females of M. mustelus captured in January and in February were examined.In the first pregnant female captured in January, totally 15 embryos (10 females and 5 males) were found.Eight of them were in the right uteri and other seven were in the left uteri.The total lengths of these embryos ranged from 23.8 to 29.3 cm (Fig. 12a).On the other hand, in the second pregnant female caught in February, totally 10 embryos (5 females and 5 males) were observed.Five of them were in the right uteri and other five were in the left uteri.The total lengths of these embryos ranged from 23.2 to 35.6 cm (Fig. 12b).

DISCUSSION
In the presently reported study, the lengths and weights of Mustelus mustelus ranged from 47.1 to 149.0 cm and from 315.6 to 8776.0 g for males and from 44.3 to 162.6 cm and from 241.3 to 12 060 g for females, respectively.The maximum lengths of the same species were reported as 165 cm for female off South Africa (Smale and Compagno 1997); 164 cm in Mossel Bay, South Africa (Goosen and Smale 1997); 165 cm in the Adriatic Sea, Mediterranean (Maddalena et al. 2001); 148.3 cm in the Adriatic (Pallaoro et al. 2005); 123 cm in the Gulf of Gabès (Saïdi et al. 2008); 122 cm in the Gulf of Gabès again (Saïdi et al. 2009).The maximum lengths reported around the Turkish coasts are given Table 3.In this study, the maximum lengths of females and males were found to be 162.6 to 149.0 cm, respectively, in the north-eastern Mediterranean.These values are the highest reported for the Turkish coasts.The vertebral band age readings showed that the female age ranged from 0 to 25; 51.51% of them were in the age group 2; while males were in the range of 0 to 20 years of age; 43.24% of them were in the age group 2. Goosen and Smale (1997) found that the age range of their females was between 0 and 24 years of age and males were between 1 and 17 years of age.Since there has been no other age study concerning this species, a comparison was made with another species of this family.For instance, a maximum age of 16 years was reported for Mustelus antarcticus Günther, 1870 (see Moulton et al. 1992) Molina et al. 2017).
The overall length and weight growth parameters, estimated based on the von Bertalanffy equation, assumed the following values: L ∞ = 195.13cm, W ∞ = 20 060 g, K = 0.06, t 0 = −4.27years.Since there were not enough samples in the study, the above-mentioned calculations were not done separately for both sexes (females and males).All these parameters calculated for different species of the same family are summarized in Table 4.
Length-weight relations may show temporal or spatial variations due to their size range, reproductive activities, and stage or environmental factors such as water temperature, food quality and availability, diseases, and competition (Wootton 1990).The lowest b value was reported to be 2.77 by Hepkafadar (unpublished * ) in the Gulf of Izmir, while the highest was given by Filiz and Mater (2002) in the northern Aegean Sea.The comparison of the total length-weight relations of Mustelus mustelus (Table 3) shows the differences between the patterns of growth between regions.The differences can be attributed to the combination of several factors such as the number and size of examined individuals.
Lengths at 50% maturity (L 50 ) were found as 109 cm at 8 years of age for females and 92 cm at 7 years of age for males.The length at sexual maturity was reported as 125-130 cm for females and 95-130 cm for males in South Africa (Smale and Compagno 1997); 59-93 cm for females and 52-57 cm for males in Mauritania (Khallahi unpublished ** ); 90-104 cm for females and 82-95 cm for males in Senegal (Capapé et al. 2006); 107.5-123.0cm for females and 88.0-112 cm for males in the Gulf of Gabès (Saïdi et al. 2008).Saïdi et al. (2008) reported that although both gonads were active in males, only right ovaries were active in females.When GSI and condition factor are evaluated together, the reproduction period of M. mustelus is estimated to take place from January to February in the north-eastern Mediterranean Sea.
These reproduction data should be taken into account for reproduction and nursery areas of M. mustelus in the north-eastern Mediterranean and decision makers dealing with fisheries management should consider establishment of marine protected areas with regional fisheries organizations.The results of this study could give useful insight for management plan and conservation of M. mustelus in the Mediterranean coast of Turkey.

FrequencyFig. 3 .
Fig. 3. Total length frequency analysis of Mustelus mustelus from the Gulf of Iskenderun, north-eastern Mediterranean Sea

Fig. 8 .
Fig. 8. Age-frequency distribution of Mustelus mustelus from the Gulf of Iskenderun, north-eastern Mediterranean Sea

n
= number of individuals studied, a = intercept, b = slope, CI = confidence limit, R 2 = coefficient of determination.

Fig. 12 .
Fig. 12. Examples of dissected pregnant females of Mustelus mustelus caught in January 2014 (A) and February 2014 (B) from the Gulf of Iskenderun, north-eastern Mediterranean Sea

Table 1
Length and weight of individual age groups of Mustelus mustelus from the Gulf of Iskenderun, north-eastern Mediterranean n = number of individuals studied.

Table 3
Comparison of total length-weight relations of Mustelus mustelus in various regions Hepkafadar unpublished = see footnote on page 34.