AGE AND GROWTH OF CAPOETA PESTAI ( ACTINOPTERYGII : CYPRINIFORMES : CYPRINIDAE ) IN A SMALL RIVER ENTERING LAKE EĞIRDIR , TURKEY

Background. Capoeta pestai (Pietschmann, 1933) is commercially important fi sh species in Turkey. The knowledge of the biology of this species is very scarce. This species has a ‘critically endangered’ status on the IUCN Red List. The objective of this study was to determine the age and growth rates of C. pestai, and to evaluate whether otolith morphometrics could be used as an age predictor. Material and methods. Fish were collected from a small river entering Lake Eğirdir between January and December 2012. Fish samples were obtained from shallow parts of the river (30–40 cm depth) seasonally in a single station by using an electrofi shing devise. Results. Annual growth increments were counted on the otoliths of 154 specimens, ranging from 6.2 to 26.8 cm total lengths. Estimated ages ranged from 0+ to 7+ years old. The von Bertalanffy growth curve was fi tted to the age/total length data as follows: L∞ = 47.83 cm (TL), K = 0.081, t0 = –1.840, for females; L∞ = 41.25 cm (TL), K = 0.086, t0 = –1.99 for males; L∞ = 52.45 cm (TL), K = 0.067, t0 = –1.84 for all specimens. No signifi cant differences in morphometric measures (length, width, and mass) were found between left and right otoliths (paired t-test, P > 0.05). Otolith length and width showed signifi cant linear relations with the total length while otolith mass represented by power model. Conclusion. This study provides the fi rst information on age, growth and otolith morphometric parameters of C. pestai. In addition, otolith morphometric measurements could possibly be used in future as an age predictor for C. pestai.

Age determination of fi sh is commonly based on counts of otolith growth rings.It is often a time-consuming and labour-intensive task (Francis andCampana 2004, Fey andLinkowski 2006).Reliable age estimates are especially diffi cult for long-lived species because of the slow growth and narrow increments in the older growth rings (Stransky et al. 2005).Moreover, accurate otolith ageing requires much practice.Therefore, many studies have used alternative methods for ageing such as the relations between fi sh size or otolith morphometric measurements and age (Francis and Campana 2004, Fey and Linkowski 2006, Steward et al. 2009, Javor et al. 2011, Matić-Skoko et al. 2011, Škeljo et al. 2012).Age and growth studies have been carried out on different Capoeta species (Türkmen et al. 2002, Alp et al. 2005, Kalkan 2008, Sen et al. 2008, Calta et al. 2010, Çoban et al. 2013, Ayyildiz et al. 2014, Emre et al. 2014, Innal 2014), however no studies have been conducted related to population dynamics for Capoeta pestai.
Therefore, the main goal of this study was to determine the age and growth rates of C. pestai, and to evaluate whether otolith morphometrics could be used as an age predictor.

MATERIALS AND METHODS
Sampling.Fish collection was carried out from a small river entering Lake Eğirdir, located at 38°03′24″N, 30°51′58″E, nearly 917 m altitude (above sea level) in the Turkish Lakes Region and is 186 km north of Antalya.The bottom of the sampling site was commonly rocky with pebbles and the water was generally clear with slow running.Fish samples were obtained from shallow parts of the river (30-40 cm depth) seasonally in a single station by using an electro shocker (Honda EU20i, 220W 12V 8A) between January and December 2012.The electrodes were fi tted with two wooden spreaders in order to keep them parallel at a distance of 50-60 cm.Length-weight relation.Total length (TL) of specimens was measured to the nearest 0.1 cm and the fi sh were weighed to the nearest 0.01 g.Sex determination was performed by direct observation of gonads.The sex ratio (number of males to each female; M : F) of the samples was analysed.The Mann-Whitney U test was applied to test the existence of signifi cant differences between sexes according to the total length.The relation between the total length and the total weight was calculated for each sex separately using a power function: where: b is the regression coeffi cient and a is the regression constant.The regression parameters a, b, and the coeffi cient of determination (r 2 ) were estimated for all individuals and for each sex with log-transformed data by linear regressions.The allometric index value (b) was compared to the theoretical value of 3 by a t-test (Zar 1984).Analyses of covariance (ANCOVA) were applied to determine any signifi cant differences in the linear relations between sexes.Age and growth.The lagenar otolith (asteriscus) pairs were removed, cleaned, dried, and stored in plastic vials.
From each pair, one otolith was randomly selected and immersed in plastic vial with glycerine solution for a week.Counts of rings in each otolith were read blind by experienced reader, without knowing the fi sh length or the date of capture.All otoliths were read twice and fi nal age estimates achieved when the same results were obtained from the two readings.A stereomicroscope with objective lenses with nominal magnifi cations ranging from ×0.5 to ×5.6 was used for the counts.
Ages were determined by counting the number of opaque and translucent rings from the nucleus to the outer edge of the otolith (Fig. 1).The von Bertalanffy growth curve was fi tted to the length at age data using non-linear least squares parameter estimation (von Bertalanffy 1938): where TL is the fi sh length at age t (year), L ∞ the theoretical asymptotic length, K the growth rate coeffi cient, and t 0 the theoretical age when fi sh length is zero.Separate analyses were carried out for males, females and all specimens.Otolith morphometrics.Otolith length (OL) and width (OW) were measured to the nearest 0.01 mm using Q Capture Imaging Software and weighed (OM) to the nearest 0.00001 g using a Shimadzu electronic balance.OL was defi ned as the longest axis between the anterior and posterior otolith edge and OW as a distance from the dorsal to the ventral edge.Differences between left and right otoliths were tested by paired t-test.The relation between the somatic growth and the otolith growth was investigated by linear regression.Relations between age and otolith weight were calculated using the power model.

RESULTS
Length-weight relation.A total of 154 individuals were collected; 44 (28.5%) were females, 104 (67.5%) males and 6 (4%) individuals undetermined sex.Female and male total lengths ranged from 9.8 to 26.8 cm and from 6.2 to 22.0 cm, respectively (Fig. 2).A higher proportion of males were observed in the ≤ 20.0 cm length classes while females were more abundant in the ≥ 20.0 cm length classes.Also, only females were sampled in the larger than 22 cm length classes.The overall male : female ratio (M : F = 2.36 : 1.00) was biased in favour of males.The Mann-Whitney U test revealed signifi cant differences between sexes, regarding TL (U = 940, z = -5.654;P < 0.05).The parameters of the length-weight relations are provided for each sex and all individuals in T able 1.Our data suggested that Capoeta pestai showed positive allometric growth.The parameter b of length-weight relations was signifi cantly different from 3 (P < 0.05).The ANCOVA test indicated that length-weight relations were not signifi cantly different in slopes or intercept between the two sexes (ANCOVA, n = 154; P > 0.05).

Table 1
Parameters of the length-weight relations for males, females and all individuals (males, females, immature) of Capoeta pestai in a small river entering Lake Eğirdir  2).Most of the fi sh, accounting for 61.7% of the total sample, were between 2+ and 4+ years old.Fish younger than 1+ year old and older than 6+ years old were poorly represented in the sample (Fig. 3).Males were dominant in the younger age groups (1+, 2+, 3+, and 4+), while the females were abundant in age classes 5+ and 6+.Moreover, age classes of 7+ years represented by only females.The von Bertalanffy growth curves for total lengths at age of females, males and all specimens are shown in Fig. 4. The estimated parameters of the equation were: L ∞ = 47.83 cm (TL), K = 0.081, t 0 = -1.840,for females; L ∞ = 41.25 cm (TL), K = 0.086, t 0 = -1.99 for males; L ∞ = 52.45cm (TL), K = 0.067, t 0 = -1.84 for all specimens.Their otolith length, width, and mass ranged within 1.16-3.91mm, 0.89-3.37mm, and 0.0003-0.0074g, respectively (Table 3).Statistically no signifi cant differences in morphometric measures (OL, OW, and OM) were found between left and right otoliths (paired t-test, P > 0.05) or between the sexes (P > 0.05).OL and OW showed signifi cant linear relations with the TL while OM represented by power model (Table 4).The relations between otolith morphometric measurements and age were shown in Fig. 5.A power model explained between 82.11% and 87.72% of the variation in age.Most precise age estimations were obtained from the OM data (r 2 ) followed by the OW and OL (Table 5).

DISCUSSION
The analysis of covariance showed no signifi cant difference between males and females in the lengthweight relation, although females tended to be slightly longer than males.The exponents of length-weight relation of the Capoeta pestai estimated in this study, shows a positive allometric growth (b = 3.117).This value is close to that obtained for the same species from the Melendiz Creek and Lake Eğirdir (Erk'akan et al. 2013, Ayyildiz et al. 2015).
The presently reported study provides the fi rst information about age, growth rates, and the otolith morphometric measurements of Capoeta pestai by using asteriscus otolith.Asteriscus otoliths are useful structures for age estimation of C. pestai.The otoliths of C. pestai showed clearly identifi able opaque and translucent bands.Considering the otolith ring formations as annual, the maximum age of C. pestai were determined as 7+.Males of the population were dominated by younger age classes, while the females were abundant in the older age classes.The fi ndings of the current study are consistent with those of Türkmen et al. (2002) who found that males were dominated at younger age classes for Capoeta umbla (Heckel, 1843).
Growth parameters calculated in this study represent the fi rst estimations of Capoeta pestai.Therefore, we were compared growth data with the different species of the genus Capoeta in Turkey (Table 6).The theoretical   the study results may be due to the specifi c method of investigation.A more reliable comparison can be reached using standardized investigation method.In this study, K values were close to 0 (zero) that consisted with other studies conducted for different Capoeta species (Türkmen et al. 2002, Alp et al. 2005, Kalkan 2008, Elp and Sen 2009).
Otolith morphometric characteristics (OL, OW and OM) examined in this study, were related to the fi sh length of Capoeta pestai.This result showed that the fi sh size could be estimated by using the otolith morphometric measurements.The best model for predicting the fi sh length of C. pestai was found for OM.Furthermore, many researchers have shown that otolith morphometric characteristics could be effectively used instead of growth ring counts (Boehlert 1985, Labropoulou and Papaconstantinou 2000, Megalofonou 2006, Matić-Skoko et al. 2011, Škeljo et al. 2012).The collection of fi sh and otolith morphometry is simpler, faster and more economical (Silva et al. 2015).With this perspective, the use of otolith morphometric measurements for age determination of C. pestai could provide an acceptable method which is economically acceptable and easy to perform compared to growth ring counts.
In this study, OL, OW, and OM increased with age throughout the life of investigated fi sh.All models provided accurate estimates, however, OM was the best predictor of fi sh age for Capoeta pestai.Many authors have emphasized the importance of OM as age predictor (Labropoulou and Papaconstantinou 2000, Megalofonou 2006, Beyer and Szedlmayer 2010, Zorica et al. 2010, Matić-Skoko et al. 2011).In addition, a recent study by Steward et al. (2009) demonstrated that only otolith weight and otolith thickness continued to increase throughout the life of the fi sh.The same authors also pointed out those otolith parameters have higher correlations with age than fi sh size parameters.Also, Boehlert (1985) reported that fi sh size and the otolith size are correlated, however, otolith size more correlated with fi sh age than the fi sh length.
In conclusion otolith morphometric measurements could possibly be used in future as an age predictor for Capoeta pestai.However, this study was limited by the relatively small sample size of some year classes.Therefore, a re-assessment should be done to better estimate the age and growth of C. pestai.

Fig. 2 .
Fig. 2. Length-frequency distribution of males, femalesand all Capoeta pestai specimens collected in a small river entering Lake Eğirdir, during the sampling period

n
= number of specimens, a = slope of the regression line, b = y-intercept, r 2 = coeffi cient of determination; TL = fi sh total length, OL = otolith length, OW = otolith width, OM = otolith mass.

Table 3
Morphometric measurements of asteriscus otolith according to the age of Capoeta pestai from a small river entering Lake Eğirdir, Turkey OL = otolith length, OW = otolith width, OM = otolith mass.

Table 4
Parameters of the relation between the otolith measurements and the fi sh total length for Capoeta pestai from a small river entering Lake Eğirdir, Turkey

Table 5
Parameters of the linear relations between the observed age and the estimated age from otolith morphometrics of Capoeta pestai collected from Lake Eğirdir during the study period