Bangladesh rich with approximately 260-265 freshwater indigenous fishes out of them 160 species considered as small indigenous species (SIS) (Hanif et al., 2015a; Rahman, 2005). These small fishes are have high nutritional value as they contain protein, vitamin A, iron, calcium, phosphorus etc. The Atrai River having total length is about 380 km (Ahmed et al., 2013) in which 61-74 species out of 265 freshwater fishes (Chaki et al., 2014; Mia et al., 2019) are available. Out of them Pethia conchonius is a self-recruiting and naturally available small indigenous species but the abundance of this species is constantly and easily facing in upheaval due to artificial and natural factors (Stoddard et al., 2006). It is also an important ornamental fish in aquarium (Rahman, 2005). Although SIS have high economic and nutritional importance, they are regularly facing to the indiscriminate exploitation of brood and young using destructive fishing gears (Hanif et al., 2015b; Islam et al., 2019; Mia et al., 2019; Siddik et al., 2014). 

The knowledge on the morphology and health status such as condition factors is primarily required for their conservation, management, or domestication (Islam and Mia, 2016; Islam et al., 2017, 2018). For example, length and weight bears a truthful tool for the measurement of growth, health, and community status (Philips, 2014; Sabrah, 2015; Sarkar et al., 2009), the stock assessment (Chaklader et al., 2015; Siddik et al., 2016), and management and conser-vation of the fisheries resources (Ilkyaz et al., 2008; Pathak et al., 2013). It also uses comprise between sexes of population and life history of fishes from different areas (Akel and Philips, 2014; Sabrah, 2015). Length-length relationship (LLR) is also useful for equivalence of length type when data are summarized (Simon and Mazlan, 2008). LLRs are essential to know the relative growth rate (Mouto-poulos and Stergiou, 2002), stock assessment and population structure of fishes in each aquatic habitat (Kara and Bayhan, 2008). Condition factors (CFs) are used for assimilating the condition, fatness, or well-being of fishes. It provides indication on phy-sical status of fishes and fish community to manage and protect of natural populations (Muchlisin et al., 2010; Sarkar et al., 2009). It can also influence the reproductive cycle or other physiological factors before high mortality rates are suffered (Nehemia et al., 2012; Victor et al., 2014). Fultons condition factor (CF_f) is a main parameter used in fishery research and have been closely related since it was first proposed (Froese, 2006). Variations of CF_f between species are greatly influenced by the body shape, which is well described by the form factor (Froese, 2006). Related body weight (BW_r) was used to recognize the prey availability, food abun-dance and gonad maturation of fishes (Anderson and Neumann, 1996).

Presently studies are available on LWRs, LLRs, and CFs was precisely texted for different freshwater fishes in Bangladesh (Islam et al., 2016; 2017). To the best of our knowledge, very few earlier reports are available on LWRs, LLRs and CFs of Pethia conchonius. Therefore, in this study, aims were considered to the know of the length-weight relationship (LWR) and length-length relationship (LLR) including the condition factors (CFf) of P. conchonius captured from the Atrai River of Dinajpur district of Bangladesh. 

Sample collection

A total of 1746 fish individuals was captured from Khansama (KS, KS, 25.937° N and 88.722° E) and Mohanpur (MP, 25.534° N and 88.762° E) stations of Atrai River at monthly interval during morning (07:00-10:00 AM) using push net (1.50 × 1.00 m², mesh size 6 mm) and seine net (15 × 3.5 m², 4 mm) with the help of commercial fishermen (Fig. 1). The collected fresh samples were conserved in ice box and immediately transferred to the laboratory of Fisheries Biology and Genetics under Hajee Moha-mmad Danesh Science and Technology University, Dinajpur, Bangladesh. Next, the five morphometric characteristics such as the total length TL, standard length SL, head length HL, fork length FL and body weight BW were measured by the method of Froese, (2006). However, TL, SL, HL, FL, and BW were the calculated with the help of slide calipers nearest to 0.1 cm for each specimen where BW (g) were taken using a digital electronic balance (HD-602ND, MEGA, Japan) to the nearest 0.1 g accuracy. Lastly, the fish samples preserved with the 10% buffered formalin for future study at the laboratory.

Fig. 1: Fish specimens collected from two sampling stations such as Khansama (KS) and Mohanpur (MP) by using push net (1.50 × 1.00 m², mesh size 6 mm) and seine net (15 × 3.5 m², 4 mm).

Calculation of length-weight relationship

Length-weight relationship (LWR) of the fishes is known as useful tools for determining biomass and to judge fish individuality from small number of specimens and to contrast health grade, plumpness, or well-being of species (Froese, 2006; Le Cren, 1951; Ndome et al., 2012). It also measures the variation from the expected weight for length of the individual specimen of fishes. However, the esti-mation for the female and male individuals of P. conchonius done separately using the conventional cubic or log-transformed formula described by Le Cren (1951) as BW = a TL^b or Log BW = log a + b log TL. Where, BW = Body weight of each sample of P. conchonius (g), TL = Total length (cm), a = Coefficient related to body form, and b = An exponent indicating isometric growth (b = 3.0) or allometric growth (3.0 < b > 3.0) followed by the Simon and Mazlan, (2008). The regression parameters “a” and “b” of the linear equation were calculated through the following the equation as a = Y‒bX and b = [n∑XY‒∑XY] / [n∑X²‒(∑X)²]. Where b = Exponent (slope), Y = Dependent variable, X = Independent variable, a = Intercept (constant), n = Number of individuals.

Calculation of length-length relationship

There exists a simple linear relationship between two linear dimensions of aquatic animals since in-creases of length measurements are proportional to each other over the period of growth progression. TL, SL, HL, FL, and BW were calculated with the help of slide calipers nearest to 0.1 cm. Relationship between two length types of P. conchonius is linear that can be measured in the form of a straight-line equation as the Y = a + bX followed by Islam et al. (2017). Where Y = Dependent variable, X = Independent variable, a = Intercept (constant), b = Exponent (slope).

Determination of condition factors

Fulton conditionv factor 

The physical condition and health status of a specimen is known as Fultons condition factor (CF_f) was determined to calculate as CF_{f =} (BW × 100) / TL³. Where CF_f = Condition factor of an individual, BW = Body weight (g), TL = Total length (cm).

Relative body weight

Relative body weight (BW_r) is mainly used for the regularly used conserving of a species (Bister et al., 2000). Therefore, it is a good physiological indicator contrasting experiential body weight (BW) with the standard body weight (BW_s) of an individual of same fishes and same length (Giannetto et al., 2012) calculated as BW_r = (BW/a TL^b) × 100 (Froese, 2006). Where BW_r = Relative body weight, BW = Body weight (g), TL = Total length (cm) and “a” and “b” = Regression parameters estimated from the LWRs.

Form factor

Form factor (a_3.0) was also used to differentiate body shape of a fish from. According to Froese, (2006) form factor (a_3.0) was estimated through an equation as a₃.₀ = 10 log a-S (b-3) (Froese, 2006). Where “a” and “b” = Regression parameters, S = -1.358 reported by Froese, (2006) to estimate a_3.0 by plotting log10 “a” vs. “b” due to lack of information on LWRs for P. conchonius.

Statistical analysis

All sorts of statistical analysis for each species were performed by using SPSS (Statistical Package for Social Science) Version 22.0 software and Microsoft Office Excel (IBM Corporation, 2013) and the PAST (Paleontological statistics, version 3.10). According to Froese, (2006), the 95% confidence interval (CI) was determined for the regression parameters “a” and “b”. To find out the significance differences of regression coefficient (b) from isometric value (b = 3) for the LWR. The respective critical values allowed the purpose of the “b” values statistically and their addition in the isometric range (b = 3) or the allometric range (3.0 < b < 3.0). More-over, one-way analysis of variance (ANOVA) tested to notice dissimilarities (P < 0.05 or 0.01) of this species based on CF_f and BW_r values followed by Tukeys pairwise post-hoc test. A Spearman rank correlation coefficient (r_s) test used to know the relationships of the condition factors (CF_f and BWr) with TL, SL and BW of P. conchonius, collected from this river. 

Very few earlier reports are the available on P. conchonius except for (Mir and Mir 2012; Gupta and Tripathi, 2017). Besides, the result of this study on P. conchonius also judged with closely related fishes. 

Length-weight relationship 

The calculated values of LWRs and regression factors using a total of 1748 fish individuals of P. conchonius from the Atrai River of Dinajpur district in Bangladesh are shown in Table 1. Here, TL of the present species were ranged from 4.2 to 8.1 cm and 4.3 to7.8 cm while BW varied from 1.02 to 8.65 & 1.09 to 7.73 g for female and male P. conchonius, respectively (Table 1). Moreover, LWRs calculated as BW = 0.0108 TL^2.764 to BW = 0.0222 TL ^3.213 and BW = 0.0057 TL^2.671 to the BW = 0.0332 TL^3.507 for female and male, respectively. Thus, the calculated a-values from LWRs were found within the expe-cted limit as 0.0032-0.0162 (Froese et al., 2014). The calculated b-values also lied within the expected limit as 2.5-3.5 or Bayesian limit as 2.88-3.26 (Froese, 2006; Froese et al., 2014). This data sugg-ested that fish growth was mostly isometric or posi-tive allometric rather than negative allometric. Gupta and Tripathi, (2017) found that length, “a” and “b” values of P. conchonius that collected from the Ganga Rivers in India ranged from 3.8 to 11.0 cm, -1.816 to -1.711 and 2.548 to 2.665, which were close to the present findings. Although different species but had similar genus and behavior, thus, LWR relationship was previously recorded as BW= 0.0139 TL^3.03   and BW= 0.0174 TL^2.88   for male and female P. ticto species (Hossain et al., 2014) and as BW= 0.004 TL^3.396 and 0.011 TL^3.966 for Puntius sophore. Rahman et al. (2012) and 0.043 TL^2.93 for P. ticto (Alam et al., 2013).

Table 1: Descriptive statistics and LWRs parameters for fish species in the Atrai River Dinajpur, Bangladesh.

KS, Khansama; MP, Mohanpur; n, number of specimens; TL, total length; BW, body weight; a*, anti-log a; a, intercept; b, slope; r2, coefficient of determination; Min, Minimum; Max, Maximum; NAG, negative allometric growth; IG, isometric growth; PAG, positive allometric growth.

Lastly, coefficient of determination (r²) also varied from 0.753 to 0.988 and 0.817 to 0.973 for female and male, respectively (Table 1); may be varied due to the differences in sample size, range of length, age, sex, season, ecology, habitats (Jobling, 2008; Khan and Sabah, 2013), and the gonadal maturity (Tarkan et al., 2006) that were not examined in the present study.

Length-length relationship

For LLRs, a relationship of TL, SL, HL and FL was considered (Table 2). Here, “a”-values ranged from 1.204-2.561 to 1.074-4.752; “b” from 0.527-1.034 to 0.473-1.098; “r²” from 0.302‒0.976 to 0.003‒0.992 for female and male P. conchonius, respectively (Table 2). Hossain et al. (2014) reported that rela-tionship between TL and SL of P. ticto, the values of “a” found as 0.315, “b” as 1.21 and “r²” as 0.978, respectively. A relationship between TL and FL, values of “a” found between 1.161 to 2.302 and 1.088 to 4.509; “b” varied from 0.197 to 0.966 and 0.549 to 1.015 “r²” between 0.310 to 0.975 and 0.006 to 0.984 for female and male, respectively. Hossain et al. (2014) investigated on LLRs of P. ticto were 3.24-6.37, 3.71-8.00 and 3.24-8.00 cm, respectively with BW were 0.43-4.16, 0.59-8.94 and 0.43-8.94 for these sexes. Besides, the regression parameter “a” was 0.0139, 0.0174, and 0.0157; “b” was 3.03, 2.88, and 2.95; and “r²” were 0.952, 0.960, and 0.954 for male, female, and combined sexes, respectively. Hossain, (2010) estimated on LLRs (TL vs. FL) for three small freshwater fishes where the values of “a”, “b” and “r²” were as 0.309, 1.129 and 0.995 for A. mola; and -0.075, 1.036 and 0.960 for P. ticto, respectively. The values of “a”, ranged from 1.123 to 1.746 and 0.961 to 4.526; “b” 0.697 to 1.007 and 0.034 to 1.113; “r²” 0.567 to 0.986 and 0.001 to 0.992 for female and male from FL vs. SL, respectively. From FL vs. SL relation-ship, the values of regression parameters (a, b, and r²) calculated as 0.196, 1.076 and 0.995 for A. mola and 0.053, 1.133 and 0.932 for P. ticto, respectively (Hossain, 2010). For HL vs. TL, “a” values ranged from 0.186 to 0.751 and 0.227 to 0.640; “b” values from 0.193 to 0.940 and 0.153 to 1.024; “r²” values from 0.016 to 0.928 and 0.036 to 0.830 for female and male, respectively. A linear relationship bet-ween HL and TL of A. mola and P. ticto was estimated where the values of “a”, “b” and “r²” recorded as -0.457, 0.235 and 0.955 for A. mola and 3323, 1519 and 0.879 for P. ticto, respectively (Alam et al., 2013). In HL vs. SL relationship, values of “a” ranged from 0.263 to 0.835 and 0.294 to 0.874, “b” from 0.111 to 0.969 and 0.158 to 0.900, r² from 0.011 to 0.923 and 0.005 to 0.860 for female and male respectfully. While “a”, “b” and “r²” were fluctuated from 0.238 to 0.832, 0.126 to 0.950 and 0.009 to 0.919 from HL vs. FL among the sexes of P. conchonius in this river that were not possible to compare due to lack of earlier findings.

Table 2: Length-length relationships of TL, SL, HL, and FL of Pethia conchonious.

KS, Khansama; MP, Mohanpur; n, number of specimens; TL, total length; SL, standard length; HL, head length; FL, fork length; a*, anti-log a; a, intercept; b, slope; r², coefficient of determination.

Condition factors

In the Atrai River, values of the Fultons condition factor (CF_f) were 1.0796‒2.104 and 1.172‒2.105 for female and male P. conchonius, respectively (Table 3). Mir and Mir, (2012) recorded that CF_f ranged from 0.57‒0.98 in P. conchonius which is lower than the present values might be due to the geogra-phical differences. Relative body weight (BW_r) ranged from 73.989 to the 126.880 and 68.602 to 126.184 for both sexes (Table 3). According to Rahman et al. (2012), BW_r ranged from 48.62 to 179.96 (102.28±16.38) in Puntius sophore from the Chalan Beel, Bangladesh. Hossain et al. (2012) also recorded the values of the CF_f were1.55 to 2.17 (1.78±0.14) whereas BW_r were 87.54 to 121.82 (100.10±7.71) of P. ticto, more or less similar ten-dency to the present findings. Moreover, a fish was scanty and elongated with lean (CF_f = 1.0), sound (CF_f = 1.20) and healthy body (CF_f = 1.40) reflected by Barnham and Baxter, (1998). So, the female (1.268±0.014 < CF_f < 1.667±0.016) and the male (1.353±0.033 < CF_f < 1.862±0.036) sexes of the P. conchonius were in lean, sound health and healthy body of the fishes collected from the Atrai River. Thin and elongated body form where the differences may be due to food loads and the sexual maturity (Gupta et al., 2011). Besides, the values of BW_r decreasing fewer than 100 for a specimen, stock or fish community represent as little prey availability or high predation whereas values above 100 designate vice-versa (Rypel and Richter, 2008).

Table 3: Descriptive statistics and one way ANOVA for condition factors (CFs) of Pethia conchonious.

So, the average values very close to 100 expressing an equilibrium relation with prey and predator (Ander-son and Neumann, 1996). The foam factor of the P. conchonius ranged from 0.0001‒0.089 and 0.0004‒0.046 for female and male, respectively (Table 3).

Hossain et al. (2012) found that the foam factors of Ailiichthys punctata were 0.0062. The values of this factor were 0.0138, 0.0345 and 0.0435 for P. sophore based on TL, FL, and the SL, respectively (Abedin et al., 2020; Rahman et al., 2012).

KS, Khansama; MP, Mohanpur; n, number of specimens; Min, Minimum; Max, Maximum.

Spearman rank correlation (rs) for the condition factor

In this study, the Spearman rank correlation (r_s) test of condition factors such as CF_f and BW_r with some morphometric characteristics of P. conchonius is shown in the Table 4. 

Here, CF_f were significantly correlated with TL (-0.220 < r_s> 0.122) in May‒August and October‒November, SL (0.073 < r_s> 0.130) in May‒July and BW (0.230 < r_s> 0.469) in all months except October and November (Table 4).

Moreover, BW_r showed significant correlations with TL (0.057 < r_s> 0.247), SL (0.074 < r_s> 0.301) and BW (0.234 < r_s> 0.462) almost all study periods in the Atrai River (Table 4).

Table 4: Spearman rank correlation coefficient (rs) for the condition factors (Fultons condition factor and relative body weight) with lengths (cm) and body weights (g) of Pethia conchonious fishes in the Atrai River, Dinajpur, Bangladesh.

KS, Khansama; MP, Mohanpur; n, number of specimens; TL, total length; SL, standard length; BW, body weight; Min, Minimum; Max, Maximum.

This data indicated that fish body weight showed more impacts on the health condition rather than its body length. Hossain et al. (2012) studied that CFf were significantly correlated with TL and BW, but no correlation recorded with SL in A. punctata. 

Pethia conchonius can play a significant role to the national fish production of Bangladesh. The body size of this species ranged from 4.2 to 8.1 cm and 1.02 to 8.65 g, respectively. A strong relationship was found in LWRs and LLRs. Here, the calculated a- or -b-values from LWRs were found within the expected limit as the 0.0032‒0.0162 or 2.5‒3.5 and Bayesian limit as 2.88‒3.26. In case of health status, CFf ranged from 1.0796 to 2.104 and 1.172 to 2.105 for female and male respectively while relative body weight (BWr) ranged from 73.989 to 126.880 and 68.602 to 126.184 for those sexes. During the study foam factor ranged from 0.0001 to 0.089 and 0.0004 to 0.046 for female and male. This small fish species is essential for pregnant and the lactating women, infants, and rural communities. Analyzing its length-weight, length-length, and conditioning factors will help assess its biology, growth, population structure, health, productivity, stocking density, ratio, spawning time, and season, as well as the fisheries management. This research will provide valuable information to conserve this species from extinction.

The authors would like to thank the students and local fishermen for their logistical assistance, as well as the Department of Fisheries Biology and Genetics at Hajee Mohammad Danesh Science and Tech-nology University in Dinajpur, Bangladesh.

The authors declare that there is no potential conflict of interest.