AGE AND GROWTH OF BASSAN BARBEL, BARBUS PECTORALIS (ACTINOPTERYGII: CYPRINIFORMES: CYPRINIDAE), UNDER CONDITIONS OF A DAM RESERVOIR

Barbus pectoralis Heckel, 1843 is widely distributed in lakes and river systems of the west-, south-, and southeast Anatolia, Syria, Iran, and Iraq (Coad 1991, Geldiay and Balık 2007). Populations of this species are commonly found in a few rivers (Gediz, Bakircay, Buyuk Menderes, Aksu, Goksu, Orontes, Euphrates, Tigris) of Turkey, in rather cool running waters. Barbus pectoralis (formerly identified as B. capito pectoralis) is one of the nine barbel species inhabiting Turkish freshwaters and it is commonly known in the country as ‘biyikli balik’. B. pectoralis can be adapted quite easily to water regime changes. It is a benthopelagic species inhabiting both lentic and lotic environments and it has an economic value since it is commercially fished in reservoirs, lakes, and rivers (Geldiay and Balık 2007). Furthermore, it is a commercially important species also in Anatolia because of its wide distribution, its body size (max 40.4 cm, 700.0 g; Çolak 1982) and its tolerance to different environmental regimes. There are several studies on the biology and ecology of B. pectoralis from several water bodies in Anatolia (Balık 1980, Çolak 1982, Ergene 1998, İkiz et al. 1998, Topkara and Balık 2004). In this paper, we provide new information on the age and growth of B. pectoralis from Kemer Reservoir. This may contribute to the species conservation and the management of its fisheries in the study area. Study area. The Kemer Reservoir is located in the city of Aydin, in west Anatolia (lat 37°32'N, long 28°32'E). The reservoir has an area of 14.75 km2 with a maximum depth of 52 m (April 2005). It was constructed on the Akcay Stream, one of the important tributary of the Buyuk Menderes River (Fig 1). The reservoir has an oligo-mezotrophic character (Özyalin 2007). The fish species such as: Acanthobrama mirabilis, Chondrostoma meandrense, Capoeta bergamae, Silurus glanis, Carassius gibelio, Lepomis gibbosus, Cyprinus carpio, Petroleuciscus smyrnaeus, and Leuciscus cephalus are also present in Kemer Reservoir.

ture, the specimens were preserved in 4% formaldehyde solution. Specimens were measured in the laboratory to the nearest 1 mm fork length (FL), and weighed to the nearest 0.01 g total weight (W). Age was determined by macroscopic examination of scales. For this, scales of each specimen (10-15 scales) were removed from the posterior part of the body, cleaned in a 10% solution of NaOH and treated according to Chugunova (1963).
Sex of mature specimens was easily determined to the naked eye, while microscopic examination was done for immature specimen. Sex ratios were tested with Chisquared analyses (χ 2 ). Fulton's equation was used to calculate the condition factor (K) for each individual: as K = 100 · W · FL -3 , where FL is the fork length [cm] and W is the body weight [g] (Le Cren 1951). The parameters a and b of the L-W relations (W = aL b ) were estimated by the least squares regression method. Confidence intervals of 95%were calculated for b to see if these were statistically different from 3. An age-length key was constructed and the parameters of the von Bertalanffy growth equation estimated by non-linear least-squares regression. The von Bertalanffy function L t = L ∞ (1 − e −K(t − to) ) was used to describe growth in size (for FL), where L t is the length of fish in cm at age t; K is the rate at which the growth curve approaches the asymptote; L ∞ is the asymptotic length in cm; and t 0 is the theoretical time at which the fish length is zero (Erkoyuncu 1995). For the growth in weight, the same function was used: where W t is the total weight and b is the power constant of the length-weight relation. Growth performance index (φ') was estimated according to Pauly and Munro (1984).

RESULTS
B. pectoralis is found in reservoir with a conductivity of 82-603 µmhos · cm −1 . Water temperature varied between 7.5 and 28.5ºC in January and August, respectively. Dissolved oxygen content ranged between 6.5 mg · L -1 (August) and 10.1 mg · L -1 (January). pH values were close to neutral or slightly alkaline. Secchi disc transparency was found to be 1.20 m in April and 4.10 m in July.
In total 206 specimen were caught during the study period, in this number 117 males (56.8%) and 89 females (43.2%). The fork length of all individuals collected ranged from 9.8 to 27.2 cm FL (mean 18.5 ± 3.8 cm FL) and weight from 13.04 to 317.32 g (mean 96.06 ± 4.19 g W) ( Table 1). Ages ranged from 1 to 6 years for both sexes. Since mesh size was large (12-45 mm), 0-group fish was not represented in the samples. According to the age-length key, second year class was dominant (31.6% ) ( Table 2).
The von Bertalanffy growth equations were computed as L ∞ = 35.7 (SE = 0.37) cm, W ∞ = 238.0 (SE = 35.42) g,      (Figs. 5, 6). The values of L ∞ and W ∞ of females was higher than those of males, but there was a no significant difference between the growth parameters of males and females (L ∞ : t = 1.587 < t 0.05, 204 = 1.97, W ∞ : t = 0.189 < t 0.05, 204 = 1.97). The growth performance index (φ') was found to be 2.32 for combined sexes and 2.29, and 2.31 for males and females, respectively.
The condition factor was calculated as 1.272-1.367 for males, and 1.337-1.423 for females.

DISCUSSION
The maximum estimated age was 6 years while Balık (1980) refers maximum age of 7 years in the Gediz River. Six and 8 years of age were reported from the Keban Reservoir in 1978Reservoir in and 1979Reservoir in , respectively (Çolak 1982. Ergene (1998) reported a maximum age of 7 years in the Goksu River, while İkiz et al. (1998) found the maximum age of 5 years from the Aksu River. In the Avsar Reservoir the maximum age was 6 years (Topkara and Balık 2004). The age groups found for Kemer Reservoir population were consistent compared to other B. pectoralis populations described by Topkara and Balık (2004) and Çolak (1982; in first year).
In this study males were dominant. A similar situation has been reported by Balık (1980) (gill nets and cast nets) and İkiz et al. (1998) (electrofishing). However, Çolak (1982) (gill nets), Ergene (1998) (gill nets and cast nets), and Topkara and Balık (2004) (gill nets) found females to be more numerous than males. There tends to be a surplus of males on the spawning grounds in some species, because the males remain there longer or because the males shed mature sperm gradually. The females usually leave the spawning grounds more rapidly, which also may be considered as an adaptation facilitating preservation of the females or more rapid recovery of gonads, which results in an increased population. Males usually predominate in the younger groups because they mature earlier but live less long (Nikolsky 1969).
The growth of bassan barbel in Kemer reservoir was isometric (b = 2.98) in contrast to similar to those inhabiting in Gediz River (b = 2.85) (Balık 1980), Aksu River (b = 2.89) (İkiz et al. 1998), Avsar reservoir (b = 2.81) (Topkara and Balık 2004), and Kemer reservoir (b = 3.14, t-test = 2.375, P < 0.05) Özcan (2008). The b values of L-W relations is known to vary according to age, maturity, and sex (Dulčić and Kraljević 1996). Geographic location and associated environmental conditions, such as seasonality (date and time of capture), stomach fullnes, disease and parasite loads (Le Cren 1951, Ricker 1975, Bagenal and Tesch 1978, Erkoyuncu 1995, can also affect the value of b. The L ∞ and W ∞ values were calculated as 34.7 cm and 225.28 g, respectively for males and 35.7 cm 238.00 g, respectively for females (Figs.5,6). The values of L ∞ and W ∞ of females was higher than those of males, but there was a no significant difference in the growth parameters between sexes (P > 0.05). Topkara and Balık (2004) reported also similar differences. The L ∞ and W ∞ values for the whole population were 35.5 cm and 232.15 g, and these values were similar to the those reported by Topkara and Balık (2004), but different from those reported by İkiz et al. (1998) (Table 3). The differences in growth between regions can be attributed to the difference in the size of the largest individual sampled in each area. Ergene (1998) and İkiz et al. (1998) report higher values for the condition coefficient (1.47-1.61 and 1.121-1.667 respectively), while Balık (1980) and    Topkara and Balık (2004) mention almost similar with the present study values (1.142-1.315 and 1.315-1.481, respectively). Variation in the condition factor of fish may be indicative of food abundance, adaptation to the environment and gonadal development (King 1995).
Our study provides some important information on the age and growth of B. pectoralis that would be useful for fishery biologist to propose adequate regulations for sustainable fishery management and conservation of this highly economic important fish species in Kemer Reservoir.