MATERIALS AND METHODS

Area of capture and sample collection: All fishes were collected early in the morning from Kompleks Lembaga Kemajuan Ikan Malaysia (LKIM), Kuantan, Pahang, Malaysia. All fishing vessels used purse seine nets to catch Indian mackerel and Fig. 1 shows the area of capture of Indian mackerel based on the information received from fisherman. Fishing vessels were equipped with icing systems and fish were kept at lower temperature to keep fresh. In this experiment, all fish samples were collected before sorting to avoid biasness on size. After collection, they were immediately preserved with ice in the ice box and transported to the laboratory. Samples were collected monthly for a period of 6 months (December to May, 2012). A total of 1064 Indian mackerel was sampled in this study.

Fig. 1:	Map showing area of Indian mackerel capture

Sample measurement: Upon arrival at the laboratory, total Length (L) and Standard Length (SL) of fishes were measured using a special measuring board with a meter rule calibrated in centimeters, Fish length was measured to the nearest centimeter. Body weight (W) of 2 decimal points was measured after blot drying with a piece of clean tissue. After recording length and weight of fish, it was dissected to collect gonad and determine the sex. All gonads were weighted with a digital balance. In some cases, the gonad was not developed. In those cases the sex of fish was unidentified. The length-weight relationship was calculated using the Equation (Pauly, 1983, 1993):

where, W is the weight of fish (g), coefficient a is the intercept in the y-axis, regression coefficient b is an exponent and L is the total length of fish (cm). The value of b indicates isometric growth when close to 3. The statistical significance level of R² was estimated and the parameters a and b were estimated by linear regression analysis based on the natural logarithms:

Additionally the coefficient of determination r² were estimated. The Fulton’s Condition factor (K) for each experimental fish has been calculated using the formula:

where, K is the Condition factor, W is the Weight of fish, L is the Total length of fish (cm).

Gonadosomatic index (GSI) was calculated using the Equation:

Statistical analysis: All regressions and correlation were statistically analyzed using SPSS (version 16.0) which was also applied to differing between months on condition factors and gonadosomatic index. Difference between monthly condition factors and gonadosomatic index were analyzed through the analysis of variance (ANOVA) and the difference between sex ratio was analyzed through the Chi-square (χ²) test.

RESULTS

Length-Weight relationship: Monthly of Length-Weight Relationships (LWR) of Indian mackerel were presented in Table 1 and Fig. 2. LWRs of Indian mackerel in all months were found to be linear. LWRs showed that the allometric coefficients vary between 2.5128 (April) and 3.0807 (May). LWRs of each month was significant (p<0.01) with all coefficients of determination (R²) values being higher than 0.72. A negative allometric growth of Indian mackerel was observed in January (W = 0.0155L^2.8839, r² = 0.8484, p<0.01) (Fig. 2b), March (W = 0.0135L^2.9298 ,r² = 0.8193, p<0.01) (Fig. 2d) and April (W = 0.0483L^2.5128, r² = 0.7261, p<0.01) (Fig. 2e). An isometric growth was observed in December (W = 0.0098L^3.0244, R² = 0.8841, p<0.01) (Fig. 2a) and February (W = 0.0113L^3.0015, R² = 0.8141, p<0.01) (Fig. 2c) while a positive allometric growth was observed only in May (W = 0.0086L^3.0807, R² = 0.8840, p<0.01) (Fig. 2f).

Condition factor: The condition factor (K) ranged from 1.0499-1.1320 (Fig. 3). Condition factor was significantly different between months (p<0.05). The significantly lowest mean K value was found in December and the highest was in February and April. The mean condition factor in February was comparatively slightly higher than in April, although they were statistically same (p>0.05). The mean condition factors in January, March and May were significantly lower than in February and April and significantly higher than in December. The mean condition factor (K_m) in relation to size class for both sexes is shown in Fig. 4. Overall higher K_m was observed in smaller fish of both male and female. Overall K_m of female was better than male.

Sex ratio: Out of the total of 1064 specimens of R. Kanagurta collected from December 2011 to May 2012, only 329 were able to determine the sex. Out of 329, 190 (57.75%) were males and 139 (42.25%) were females.

Table 1:	Monthly descriptive statistics and estimated parameters of length-weight relationship of Indian mackerel from December 2011 to May 2012
*Indicates significant at p<0.01

Table 2:	Reported a and b values of different fish in different locations

Fig. 2:	Relationship between total length (cm) and body weight (g) in (a) December 2011, (b) January 2012, (c) February 2012, (d) March 2012, (e) April 2012 and (f) May 2012

Overall the sex ratio differed significantly (χ² = 7.91; p<0.01). Male was significantly more than female in the Indian mackerel population in the Kuantan coastal area.

Gonadosomatic index: Monthly changes of mean gonadosomatic index (GSI) of Indian mackerel are presented in Fig. 5. The lowest GSI of Indian mackerel was observed in December and January. GSI of Indian mackerel in February was higher than March and April and followed by May. GSI rapidly increased after January and reached at peak in February (1.1958%). After February GSI was declining again. A positive relationship was observed between GSI and condition factor of Indian mackerel (Fig. 6).

Fig. 3:	Monthly mean (±95% confidence intervals) condition factor. Mean with no letter in common differ significantly (p<0.05)

Fig. 4:	Mean condition factor (Km) per length class (total length) for both sexes

DISCUSSION

Length-weight relationship is very useful in fisheries science for both applied and basic use to (1) estimate weight from length observations because direct weight measurements can be time consuming in the field (Beyer, 1987; Martin-Smith, 1996; Sinovcic et al., 2004); (2) calculate, growth, biomass and production of a population (Le Cren, 1951; Pauly, 1983) and (3) compare the life history of fishes of different localities (Petrakis and Stergion, 1995); (4) determine the relative condition of small fish compared to large fish and (5) set yield equations for estimating number of fish landed and compare the population in space and time (Beverton and Holt, 1957). Besides these, length-weight relationships allow conversion of length-growth equations to weight-growth equivalents in yield-per-recruit and related models.

The exact relationship between length and weight differs among species of fish according to their inherited body shape and within a species according to the condition (robustness) of individual fish (Schneider et al., 2000). So the equation derived from the Equation logW-loga+blogL differs every time. In the present study, the coefficient b of length-weight relationship was ranged between 2.5128-3.0807 which was in the acceptable range (Bagenal and Tesch, 1978).

Fig. 5:	Monthly changes of mean (±95% confidence intervals) gonadosomatic index (GSI) of Indian mackerel in the coastal water of Kuantan. Mean with no letter in common differ significantly (p<0.05)

Fig. 6:	Relationship between GSI and Condition Factor

The b value of most fish is around 3 (Table 2) but it can vary from 2-4 (Bagenal and Tesch, 1978). There is no previous study in the Kuantan coastal water comparing the b value of Indian mackerel. However, Sivadas et al. (2006) reported 3.38 as the b value of Indian mackerel in Calicut, India (Table 2).

In this study, the b-value of the LWR in April was very low (2.5128) compared to other month. However, this might be influenced by environmental or habitat factors. For example, differences in the water temperature, availability of food, etc. are known to influence the growth. According to Tesch (1971) length-weigh relationship of fish are affected by many factors including season, habitat, gonad maturity, sex, diet, health and preservation techniques. In this study, environmental or habitat factors were not analysed. However, more research is needed including analyzing environmental or habitat factors to understand the cause of low b value in April in Kuantan coastal water.

Condition factor is a quantitative parameter that indicates the state of the fish (fatness, maturity and spawning gonadal development and general well-being of the fish) and determine present and future population success by influencing growth, reproduction and survival (Wootton, 1990). Condition factor shows the population’s condition (welfare) during the various stages of the life cycle. The condition factor normally decreases at the start of the spawning period due to very high metabolic rates. According to Mde and Ambrosio (2002) normally condition factor increases during the reproductive period and normalization occurs immediately after spawning. However, the study of the condition factor is very important for understanding the life cycle including spawning season of fish species and contributes to adequate management. In the present study, the condition factor of Indian mackerel was lower in December and January. Condition factor of Indian mackerel in February, March and April were higher than the other months. Condition factor of Indian mackerel in May was still higher than December and January. In the present study, it was also observed that the condition factor was positively correlated with gonadosomatic index of Indian mackerel. Based on condition factor and gonadosomatic index, it can be concluded that the peak spawning season of Indian mackerel in the Kuantan coastal water is from February to April. However, the spawning season of Indian mackerel in Kuantan coastal water may fall within the period between end of January and end of May.

CONCLUSION

This study provided a basic information on the length-weight relationship and condition factor of R. Kanagurta that would be useful for fisheries management. This data can be specifically used to impose adequate regulations for sustainable fishery management in the Kuantan coastal water, Malaysia. In this study, the b-value of the length-weight relationship of Indian mackerel in April was very low compared to other month. This might be influenced by environmental or habitat factors. Therefore, more research is needed including analyzing environmental or habitat factors to understand the cause of low b value in April in Kuantan coastal water. The peak spawning season of Indian mackerel in the Kuantan coastal water was from February to April. The spawning season of Indian mackerel in the Kuantan coastal water may fall within the period between the end of January and end of May.