Introduction
Beetal is an indigenous goat breed of India and its population is approximately 0.71 million out of the 148.88 million goat population in the country (Livestock census, 2019). It is a large-sized breed with good milking ability and mainly preferred to be reared for meat and milk production. The body characteristics included coat colour as predominantly black with white patches, long and flat ears, and a large and well developed udder. The birth weight and weaning weight of this breed are approximately 2.5–3 and 10–15 kg, respectively (Khan et al., Reference Khan, Ali, Hyder and Chatta2007; Magotra et al., Reference Magotra, Bangar, Chauhan, Malik and Malik2021). The mature weights in male and female goats were reported as approximately 30 and 37 kg, respectively (Magotra et al., Reference Magotra, Bangar and Yadav2020). The genetic evaluation of growth parameters was performed earlier and the heritability was estimated as 0.06, 0.27, 0.37, 0.17 and 0.10 for birth weight, and weights at 3, 6, 9 and 12 months of age, respectively (Magotra et al., Reference Magotra, Bangar, Chauhan, Malik and Malik2021). The ages at first kidding (AFK) for a farmer’s flock and under farm conditions were 19 and 26 months, respectively, and twining rate for this breed was approximately 50% (Afzal et al., Reference Afzal, Javed and Shafiq2004; Maroof et al., Reference Maroof, Singh, Sadana, Alam and Chahal2007; Food and Agriculture Organization, 2020).
The genetic evaluation of growth traits of kids provides the basis for selection strategies to enhance genetic quality and performance of the flock (Bangar et al., Reference Bangar, Magotra and Yadav2020). Along with growth performance, reproductive efficiency in goat production systems is also one of the most important factors that influences farm profitability (Song et al., Reference Song, Jo and Sol2006; Kosgey and Okeyo, Reference Kosgey and Okeyo2007). Among the various measurements of reproduction efficiency cited in the published literature, the most commonly used measures are age at first kidding (AFK), litter size and litter weight that may provide alternatives for selection basis or may be used along with existing selection criteria (Greyling, Reference Greyling2000; Mellado et al., Reference Mellado, Valdéz, García, López and Rodríguez2006; Menezes et al., Reference Menezes, Sousa, Cavalcanti-Filho and Gama2016). Assessment of these traits under standard protocols is essential to bring genetic improvement in reproductive efficiency in goat production systems (Notter, Reference Notter2012). Additionally, early detection of highly performing does during their first kidding can be useful to increase farm profitability (Mellado et al., Reference Mellado, Mellado, García and López2005). The factors influencing doe efficiency and the source of genetic variation at first kidding might be useful for establishing potential breeding programmes in advance to achieve desirable genetic progress.
Several authors have evaluated early reproduction traits such as AFK, litter size and litter weights to estimate the possible source of genetic variation in different goat breeds (Rashidi et al., Reference Rashidi, Bishop and Matika2011; Kebede et al., Reference Kebede, Haile, Dadi and Alemu2012; Mohammadi et al., Reference Mohammadi, Moradi Shahrebabak and Moradi Shahrebabak2012; Kasap et al., Reference Kasap, Mioc, Skorput, Pavic and Antunovic2013; Mokhtari et al., Reference Mokhtari, Asadi Fozi, Gutiérrez and Notter2019). However, the low level of genetic variability among these traits has indicated non-suitability of these traits for selection programmes (Mellado et al., Reference Mellado, Mellado, García and López2005). The influence of environmental factors such as feeding, rearing and seasonal variations might be responsible for the reproductive efficiency of goats (Song et al., Reference Song, Jo and Sol2006; Moaeen-ud-Din et al., Reference Moaeen-ud-Din, Yanf, Chen, Zhang, Xiao, Wen and Dai2007; López-Sebastián et al., Reference López-Sebastián, Coloma, Tolédano and Santiago-Moreno2014). Therefore, it is of utmost importance to evaluate genetic and non-genetic factors associated with reproductive efficiency to determine the forces that need to be applied to increase the productivity and economic viability of goat production systems. To the best of our knowledge, there have been no reports on estimates of genetic parameters for early reproduction traits of Beetal goat so far.
Keeping under consideration the importance of early reproduction traits for updating an efficient breeding plan, the aim of the present study was to estimate genetic parameters of these traits in Beetal goat, while adjusting the effects of non-genetic factors.
Materials and methods
Study area
Beetal goats used in the present study were maintained at a goat breeding farm, Department of Animal Genetics and Breeding, LUVAS, Hisar (Haryana), India. The management practices at the farm were mostly uniform but some practices, such as extra bedding and grazing time, were subject to weather conditions. The common practices included 6–8 h grazing around the farm area, 300–400 g of concentrate feeding and feeding of green (clover, maize and sorghum) and dry (wheat bran) fodder. Additional rations prior to mating and during advanced pregnancy were given to animals during the study period.
The reproductive management of this breed at farm included seasonal mating, flushing, care of pregnant females and new born kids, and nutritional practices required for the herd. The deworming and flushing (extra 100g concentrate) were done prior to mating of the animals. The seasonal mating with hand mating was reported for animals during the month of May for each year. The age of first breeding was almost 18 months from their birth. Heat identification in females was reported using an aproned buck and the age of buck for mating ranged from 1.5 to 5 years. The ratio of male to female was reported as 1:20. Extra concentrate feeding during advanced pregnancy and after kidding was reported at the farm. The respective kidding occurred in the month of October. The period of weaning of kids from their dams was 3 months of age.
Data records
The dataset including 223 female kids born to 25 sires and 122 dams including details of year of birth, type of birth and dam’s weight at kidding were compiled from birth and weight register for the period from 2004 to 2019. The early reproduction traits were generated from the first kidding records of 223 females. The AFK for female goats was considered in months. The litter traits were generated depending upon size and weight at different ages. The litter size at birth per doe at first kidding (LSB) was coded as 1, 2 and 3. Similarly, litter size at weaning per doe at first kidding (LSW) was coded 1, 2 and 3. The does whose kids did not reach weaning stage were excluded for LSW trait. Furthermore, total litter weight at birth of kids born (LWB) and at weaning of kids born (LWW) were generated accordingly.
Statistical analysis
The effects of non-genetic factors such as period of birth (2004–2008, 2008–2012, 2012–2016 and 2016–2019), type of birth (single and multiple) and dam’s weight at kidding (< 28, 28–32 and ≥32 kg) on early reproduction traits were assessed using a general linear model (SPSS 20) as follows:
$Y_{ijkl} = {\mu} + P_{i} + T_{j} + W_{k}+ E_{ijkl}$
where Yijkl represents the observation of a particular trait, μ is the grand mean, Pi is the effect of ith period of birth (i = 1–4), Tj is the effect of jth type of birth (j = 1, 2), Wk is the effect of kth dam’s weight at kidding (k = 1–3) and Eijkl is random error distributed normally with mean 0 and variance σ2. The significant (P < 0.05) effects were further subjected for partitioning total variation in the source of additive and residual variations among the traits. Variance component estimation was undertaken using dyadic mixed modelling under the ‘dmm’ package in R (R v.4.0.1; http://www.r-project.org/). The dyadic mixed model provides pedigree-based partitioning of individual variation into a range of environmental and genetic variance components. The following univariate animal model was used to derive VarG(Ia) and VarE(I) variance components from individual variance in this study:
$Y = X \beta + Z_{a}a+\epsilon$
where, Y is a vector of observation of each trait; β, a and ϵ are vectors of observations, fixed, direct additive genetic, and residual effects, respectively; and X and Z a are the design matrices of fixed effects and direct additive genetic effects. As the number of records for studied traits was almost same, genetic and phenotypic correlations among studied traits were estimated under a multivariate approach using the ‘dmm’ package in R.
Results
The data structure along with descriptive statistics for early reproduction traits in Beetal goat are given in Table 1. The least-squares means for LSB, LSW, LWB, LWW and AFK were 1.27 ± 0.03, 1.25 ± 0.03, 3.24 ± 0.07 kg, 13.08 ± 0.30 kg and 27.56 ± 0.58 months, respectively. The data record for LSB was 223 does, which was reduced for LSW as 14 kids did not survive up to the weaning period. Out of 223 does kidding for the first time, 163, 59 and one females had single (73.09%), twin (26.46%) and triplet (0.45%) births. The coefficient of variation for these traits ranged from 29.76 to 35.86. The least-squares estimates for studied traits and adjusting for various non-genetic factors are presented in Table 2. The period of birth had significant (P < 0.05) influence on all traits under study. However, type of birth and dam’s weight at kidding were non-significantly (P > 0.05) associated with studied traits.
Table 1. Descriptive statistics for early reproductive traits in Beetal goat
AFK: Age at first kidding (in months); CV: Coefficient of variation; LSB: Total kids born at first kidding; LSW: Total kids weaned at first kidding; LWB: Total litter weight at first kidding; LWW: Total litter weight of weaned kids at first kidding.
Table 2. Effect of non-genetic factors on early reproduction traits in Beetal goat
Means (±SE) with different superscripts differ significantly (P < 0.05) in different levels of particular factor.
AFK: Age at first kidding (in months); LSB: Total kids born at first kidding; LSW: Total kids weaned at first kidding; LWB: Total litter weight at first kidding; LWW: Total litter weight of weaned kids at first kidding.
The estimates of variance components and direct additive heritability for studied traits are given in Table 3. The estimates of phenotypic variance for LSB, LSW, LWB, LWW and AFK were 0.21, 0.20, 0.90, 17.70 and 73.60, respectively and corresponding proportions due to additive effects were as low as 0.02, 0.01, 0.09, 0.61 and 4.23, respectively. This variance partitioning due to additive effects showed that sources of variation among these traits were due to residual components. The direct heritability estimates for LSB, LSW, LWB, LWW and AFK were found to be as low as 0.08, 0.03, 0.10, 0.03 and 0.06, respectively.
Table 3. Variance components and direct heritability estimates for early reproduction traits in Beetal goat
AFK: Age at first kidding (in months); LSB: Total kids born at first kidding; LSW: Total kids weaned at first kidding; LWB: Total litter weight at first kidding; LWW: Total litter weight of weaned kids at first kidding.
Genetic and phenotypic relationships among the studied traits were estimated under multivariate analysis and the results are shown in Table 4. The genetic correlation of LSB was positive and moderate to high (0.24–0.80) with other reproduction traits under study, except AFK (rg = −0.46). Additionally, LWB was also highly and positively associated with LSW (rg = 0.69) and LWW (rg = 0.42). However, genetic correlation of AFK was highly negative with other traits and ranged from −0.97 to −0.46. The phenotypic correlations among LSB, LSW, LWB and LWW under study were highly positive and ranged from 0.70 (LWB–LWW) to 0.87 (LSB–LSW). For AFK, it was low and ranged from 0.13 (LWW) to 0.25 (LWB). Furthermore, the residual correlations among litter traits ranged from 0.71 (LWB–LWW) to 0.91 (LSB–LSW). For AFK, it was positive but ranged low to medium with other traits.
Table 4. Genetic (above diagonal), phenotypic (below diagonal) and residual (in parenthesis) correlation estimates among early reproduction traits in Beetal goat
AFK: Age at first kidding (in months); LSB: Total kids born at first kidding; LSW: Total kids weaned at first kidding; LWB: Total litter weight at first kidding; LWW: Total litter weight of weaned kids at first kidding; NE: Non-estimable.
Discussion
We evaluated early reproductive traits in Beetal goat for effects of non-genetic and genetic factors on litter traits and AFK. As previous published literature contained few studies on this area, therefore we believed that this work may provide insight to the possible potential of early reproduction traits for improving overall goat production systems. Additionally, no previous study has been undertaken on genetic evaluation of reproductive traits in Beetal goat.
Litter size at birth (LSB) and weaning (LSW)
The LSB and LWW values from our study were close to values reported by Rashidi et al. (Reference Rashidi, Bishop and Matika2011) in Markhoz goat (LSB = 1.30, LSW = 1.20). Similarly, Bagnicka et al. (Reference Bagnicka, Wallin, Łukaszewicz and Ådnøy2007) also reported a similar value to the present study for LSB in Norwegian (1.23) goat. Our estimates were higher than findings by Mohammadi et al. (Reference Mohammadi, Moradi Shahrebabak and Moradi Shahrebabak2012) in Raeini goat (LSB = 1.05, LSW = 0.96) and Mokhtari et al. (Reference Mokhtari, Asadi Fozi, Gutiérrez and Notter2019) in Raeini Cashmere goat (LSB = 1.10, LSW = 0.80). Compared with this study, higher estimates were reported by Song et al. (Reference Song, Jo and Sol2006) in Korean native (LSB = 1.78, LSW = 1.52), Bagnicka et al. (Reference Bagnicka, Wallin, Łukaszewicz and Ådnøy2007) in Polish (LSB = 1.51), Zhang et al. (Reference Zhang, Chen, Li, Xu, Zhang and Yang2009) (LSB = 1.76, LSW = 1.62) in Boer goat, Kebede et al. (Reference Kebede, Haile, Dadi and Alemu2012) in Arsi-Bale (LSB = 1.60, LSW = 1.37), and Menezes et al. (Reference Menezes, Sousa, Cavalcanti-Filho and Gama2016) in Boer goat (LSB = 1.71). These discrepancies for litter size trait might be due to climate region and type of breeds (dairy/meat). The litter size in goat was lower in tropical and subtropical regions (˜1.3) than temperate regions (>1.6) (Devendra Reference Devendra1980; Moaeen-ud-Din et al., Reference Moaeen-ud-Din, Yanf, Chen, Zhang, Xiao, Wen and Dai2007). The home tract of Beetal goat falls under the subtropical region where fluctuation in seasonality for reproduction might be responsible for low litter size and prolificacy for this breed.
Litter weight at birth (LWB) and weaning (LWW)
The estimates for LWB from our study were in line with values given by Rashidi et al. (Reference Rashidi, Bishop and Matika2011) in Markhoz (3.50) and Kebede et al. (Reference Kebede, Haile, Dadi and Alemu2012) in Arsi-Bale (3.70). Compared with our estimates, Menezes et al. (Reference Menezes, Sousa, Cavalcanti-Filho and Gama2016) reported higher estimates for LWB (5.80) and weaning (23.42) in Boer goats. Our estimate of LWW was in accordance with the estimate given by Maghsoudi et al. (Reference Maghsoudi, Torshizi and Jahanshahi2009) in Iranian Cashmere goats (12.10). However, Rashidi et al. (Reference Rashidi, Bishop and Matika2011) also reported higher LWW values (21.90) in Markhoz goat than the present findings. A lower estimate than our study was reported by Mohammadi et al. (Reference Mohammadi, Moradi Shahrebabak and Moradi Shahrebabak2012) in Raeini goat (LSB = 2.39) and Mokhtari et al. (Reference Mokhtari, Asadi Fozi, Gutiérrez and Notter2019) in Raeini Cashmere goat (LWB = 2.49, LWW = 11.30).
Age at first kidding
AFK (in months) estimated under the current study was higher than that found by Alexandre et al. (Reference Alexandre, Aumont, Mainaud, Fleury and Naves1999) in Creole (17.20), Bagnicka et al. (Reference Bagnicka, Wallin, Łukaszewicz and Ådnøy2007) in Polish (13.89) and Norwegian (13.63), Torres-Vázquez et al. (Reference Torres-Vázquez, Valencia-Posadas, Castillo-Juárez and Montaldo2009) in Saanen (16.59), Castañeda-Bustos et al. (Reference Castañeda-Bustos, Montaldo, Torres-Hernández, Pérez-Elizalde, Valencia-Posadas, Hernández-Mendo and Shepard2014) in US dairy goat (16.9) and Mokhtari et al. (Reference Mokhtari, Asadi Fozi, Gutiérrez and Notter2019) in Raeini Cashmere (19.66). The higher estimate of AFK in our resource population might be due to breed differences (Gill and Dev, Reference Gill and Dev1972). Our estimate was close to AFK values of some Indian goat breeds reported by Kumar et al. (Reference Kumar, Kumar, Mishra and Singh2006) in Kutchi (23.28), Kumar et al. (Reference Kumar, Nagda and Sharma2012) in Sirohi goat (22.10) and Gautam et al. (Reference Gautam, Waiz and Nagda2018) in Sirohi goat (21.90).
Effect of non-genetic factors
The significant effects of period of birth on targeted traits of goat was also reported by Zhang et al. (Reference Zhang, Chen, Li, Xu, Zhang and Yang2009) in Boer, Kebede et al. (Reference Kebede, Haile, Dadi and Alemu2012) in Arsi-Bale, Mohammadi et al. (Reference Mohammadi, Moradi Shahrebabak and Moradi Shahrebabak2012) in Raeini and Mokhtari et al. (Reference Mokhtari, Asadi Fozi, Gutiérrez and Notter2019) in Raeini Cashmere goat. Gautam et al. (Reference Gautam, Waiz and Nagda2018) also reported significant effects of birth period on AFK in the Sirohi breed of Indian goat. The periodic variation in feeding, rearing and health managerial practices might have resulted in significant changes in targeted traits over periods of birth. Also, indirect effects of selection strategies used in recent years at farms on litter traits might be responsible for periodic changes in targeted traits. The non-significant effects of type of birth and doe weight at kidding on studied traits were contrary to reports by Kebede et al. (Reference Kebede, Haile, Dadi and Alemu2012). The type of birth and doe’s weight at kidding were mostly associated with the initial growth performance of the kid, which might have a negligible influence on reproduction traits. As we studied only traits at first parity, the role of parity as a fixed effect for these traits was not evaluated in this study. Additionally, season of breeding was not included in the model due to seasonal breeding (month of May each year) among the studied population.
Heritability estimates
The low-level estimates of direct heritability of litter size at birth and weaning in our study were in agreement with reports by Rashidi et al. (Reference Rashidi, Bishop and Matika2011) in Markhoz (LSB = 0.01, LSW = 0.01), Mohammadi et al. (Reference Mohammadi, Moradi Shahrebabak and Moradi Shahrebabak2012) in Raeini goat (LSB = 0.04, LSW = 0.09), Kasap et al. (Reference Kasap, Mioc, Skorput, Pavic and Antunovic2013) in Saanen (LSB = 0.07) and Menezes et al. (Reference Menezes, Sousa, Cavalcanti-Filho and Gama2016) in Boer goat (LSB = 0.00). But our estimates were lower than the reports by Bagnicka et al. (Reference Bagnicka, Wallin, Łukaszewicz and Ådnøy2007) in Polish (0.14) and Norwegian (0.18), Zhang et al. (Reference Zhang, Chen, Li, Xu, Zhang and Yang2009) in Boer (LSB = 0.12, LSW = 0.10), Kebede et al. (Reference Kebede, Haile, Dadi and Alemu2012) in Arsi-Bale (LSB = 0.14, LSW = 0.16) and Mokhtari et al. (Reference Mokhtari, Asadi Fozi, Gutiérrez and Notter2019) in Raeini Cashmere goat (LSB = 0.12, LSW = 0.23). LWB and weaning trait also had lower additive genetic variability in our resource population.
The direct heritability estimate for LWB was close to the estimate given by Zhang et al. (Reference Zhang, Chen, Li, Xu, Zhang and Yang2009) in Boer goat (0.14). However, it was higher than findings of Rashidi et al. (Reference Rashidi, Bishop and Matika2011) in Markhoz (0.02), Kasap et al. (Reference Kasap, Mioc, Skorput, Pavic and Antunovic2013) in Saanen (0.04) and Menezes et al. (Reference Menezes, Sousa, Cavalcanti-Filho and Gama2016) in Boer goat (0.01). It was lower than the reports by Mohammadi et al. (Reference Mohammadi, Moradi Shahrebabak and Moradi Shahrebabak2012) in Raeini goat (0.16), Kebede et al. (Reference Kebede, Haile, Dadi and Alemu2012) in Arsi-Bale goat (0.15) and Mokhtari et al. (Reference Mokhtari, Asadi Fozi, Gutiérrez and Notter2019) in Raeini Cashmere goat (0.17). Our estimate of direct heritability for LWW was close to the estimate by Rashidi et al. (Reference Rashidi, Bishop and Matika2011) in Markhoz goat (0.03) and Maghsoudi et al. (Reference Maghsoudi, Torshizi and Jahanshahi2009) in Iranian Cashmere goats (0.07), but it was lower than that by Kebede et al. (Reference Kebede, Haile, Dadi and Alemu2012) in Arsi-Bale goat (0.12) and Mokhtari et al. (Reference Mokhtari, Asadi Fozi, Gutiérrez and Notter2019) in Raeini Cashmere goat (0.15). These low estimates from our study for litter size and litter weight at different ages indicated that the sources of variation among these traits were mainly residual effects rather than genetic effects. It was also suggested that only modest genetic improvement could be possible through selection of these traits.
The estimate of direct heritability for AFK was lower than the estimate reported by Bagnicka et al. (Reference Bagnicka, Wallin, Łukaszewicz and Ådnøy2007) in Polish (0.13), Torres-Vázquez et al. (Reference Torres-Vázquez, Valencia-Posadas, Castillo-Juárez and Montaldo2009) in Saanen (0.31), Kebede et al. (Reference Kebede, Haile, Dadi and Alemu2012) in Arsi-Bale goat (0.25) and Mokhtari et al. (Reference Mokhtari, Asadi Fozi, Gutiérrez and Notter2019) in Raeini Cashmere goat (0.46). Castañeda-Bustos et al. (Reference Castañeda-Bustos, Montaldo, Torres-Hernández, Pérez-Elizalde, Valencia-Posadas, Hernández-Mendo and Shepard2014) reported 0.16 as the direct heritability estimate for AFK in US dairy goat. The mean value of AFK in our resource population was quite high and that might be due to the extended role of environmental effects to a greater extent resulting in low genetic variability and low estimates of heritability for this trait in Beetal goat.
Correlation estimates
Genetic correlation provides the basis for improvement of desired traits by selecting particular traits. The strong genetic association among traits indicated the possible scope for simultaneous improvement of both traits. The estimates of genetic correlation among litter size and litter weight at different ages under this study were moderate and positive, and were in accordance with reports by Rashidi et al. (Reference Rashidi, Bishop and Matika2011) in Markhoz (0.75 to 0.95), Kebede et al. (Reference Kebede, Haile, Dadi and Alemu2012) in Arsi-Bale (0.71 to 0.99) and Mokhtari et al. (Reference Mokhtari, Asadi Fozi, Gutiérrez and Notter2019) in Raeini Cashmere goat (0.12 to 0.80). This finding indicated the importance of LWB for weaning performance of kids. The negative genetic correlation between AFK and litter traits was also reported by Mokhtari et al. (Reference Mokhtari, Asadi Fozi, Gutiérrez and Notter2019) in Raeini Cashmere goat. Early kidding was associated with better lifetime production. Low AFK associated with high litter size in our study might be due to the high AFK present in the Beetal breed. Phenotypic correlation estimates among targeted traits under our study were moderate among litter traits and low for AFK combinations, which were in line with estimates reported by Rashidi et al. (Reference Rashidi, Bishop and Matika2011) in Markhoz goat.
Conclusion
The early reproduction traits in goats have importance for identifying highly performing does at initial stages to enhance farm production in terms of litter size and weight. Although effects of some non-genetic factors were non-significant for these traits in our study, the identification of periodic change must be taken into consideration. The low level of genetic variation among litter traits and AFK indicated a modest scope for genetic improvement through selection. It was suggested that the adoption of better managerial practices including feeding, rearing and health practices subjected to seasonal fluctuations may increase the reproduction efficiency of Beetal goat.
Competing interests
The authors declare no competing interests regarding publication of this paper.
Acknowledgements
The authors are thankful to the Vice-Chancellor, LUVAS, Hisar (Haryana), India for providing funding and the necessary facility to conduct this study.
