Relationship of body measurements with body weight at different ages in local (Dharuri) sheep under field conditions in Beed district of Maharashtra

Introduction

Sheep rearing is an important part of the livestock-based rural economy of Maharashtra, particularly in semiarid areas where crop production is constrained by low rainfall, seasonal feed scarcity and small landholdings. Local sheep populations, having evolved under the prevailing environmental and management conditions, are able to produce under extensive and semi-extensive systems. The local sheep of Beed district, known as Dharuri, are maintained mainly under field conditions by sheep-rearing households, but information on their growth and on the relationship between their body measurements and body weight is limited. Body weight is a key indicator of growth and productivity in sheep. It determines market price and is needed for monitoring nutritional status, selecting replacement stock and calculating doses of veterinary medicines. Weighing balances, however, are rarely available to farmers under field conditions, and easily recorded linear body measurements offer a practical alternative for estimating live weight. Prediction equations based on body measurements have been developed for several Indian sheep breeds, including Nellore (Rani et al., 2016) [1] , Harnali (Kumar et al., 2018) [2] , Malpura (Kumar et al., 2021) [3] and Gurez (Rather et al., 2022) [4] , as well as for sheep in other countries (Abbas et al., 2021; Ibrahim et al., 2021; Mekonnen et al., 2024) [5, 6, 7] . Chest (heart) girth has frequently been identified as the best single predictor of body weight because it reflects the development of the thoracic region, musculature and overall body mass. In Harnali sheep, Kumar et al. (2018) [2] concluded that heart girth was the most important trait for estimating live weight (R 2 = 0.87), and Ravimurugan et al. (2013) [8] identified chest girth as the strongest single predictor in Kilakarsal sheep. A recent meta-analysis also reported that heart girth was more closely correlated with body weight (r = 0.83) than body length (r = 0.69) or height at withers (r = 0.64) (Firdaus et al., 2025) [9] . In Deccani sheep, Narsimham et al. (2003) [10] reported positive associations between body weight and biometric measurements. The present study was undertaken to estimate the phenotypic correlations between body measurements and body weight at different ages in local (Dharuri) sheep of Beed district, and to develop age-and sex-specific regression equations for predicting body weight under field conditions.

Materials and Methods

Study area and animals

The study was conducted in Beed district of Maharashtra, in the semiarid Marathwada region of the Deccan plateau. Four sheep-rearing blocks, namely Adas, Wadhwani, Dharur and Neknoor, were selected, and sheep maintained under farmers' extensive or semi-extensive grazing systems were recorded. A total of 1,487 local sheep (484 males and 1,003 females) were studied: 348 from Adas, 392 from Wadhwani, 428 from Dharur and 319 from Neknoor. The animals were grouped into five age classes (Table 1). The marked fall in the number of males in the older age classes reflects the practice of selling male lambs at a young age while retaining females for breeding.

Body measurements

Height at withers (HW), chest girth (CG) and body length (BL) were measured to the nearest 0.1 cm with a flexible tape, with the animal standing squarely on level ground, generally in the morning before grazing. Body weight (BW) was recorded with a 100 kg spring balance wherever the owner permitted weighing. Where direct weighing was not possible, body weight was estimated from age-specific regression equations based on body measurements.

Statistical analysis

All analyses were carried out separately within each age class. Phenotypic (Pearson product-moment) correlations among BW, HW, CG and BL were computed, and their significance was tested against tabulated values of r at (n -2) degrees of freedom (Steel et al., 1960) [11] . Simple linear regressions of body weight on each measurement were fitted as 𝑌 = 𝑎 + 𝑏𝑋, where Y is body weight (kg), X is the body measurement (cm), a is the intercept and b the regression coefficient. Multiple regressions were fitted as Y = a + b 1 X 1 + b 2 X 2 + b 3 X 3 + e, where X 1 , X 2 and X 3 are HW, CG and BL, respectively, b 1 to b 3 the partial regression coefficients and e the residual. The significance of regression coefficients was tested by the t-test and goodness of fit was assessed by the coefficient of determination (R 2

). Simple and multiple regressions were also fitted separately for males and females. Homogeneity of the regression slopes of males and females was tested by fitting the model 𝐵𝑊 = 𝑎 + 𝑏𝑋 + 𝑐 ⋅ Sex + 𝑑 ( 𝑋 × Sex ) and testing the significance of the interaction term (d) by an F-test with 1 and (n -4) degrees of freedom; a non-significant F indicates that a common slope adequately describes both sexes.

Results and Discussion

Phenotypic correlations

Phenotypic correlations among body weight and body measurements are presented in Table 2. All correlations were positive and highly significant (P < 0.01) at all ages. Correlations of body weight with the three measurements ranged from 0.93 to 0.99. Chest girth had the highest correlation with body weight at 0-3 (0.98), 7-12 (0.98) and above 36 months (0.99); height at withers had the highest correlation at 4-6 months (0.97); and height and body length were equally correlated with body weight at 13-36 months (0.97). The strong positive relationships indicate that increases in body size were closely matched by increases in body weight, and support the use of chest girth as a field measurement. [8] and Harnali sheep (Kumar et al., 2018) [2] . The magnitude of the correlations in the present study (0.93-0.99), however, is considerably higher than that usually reported for sheep under field conditions; for example, Kumar et al. (2018) [2] reported phenotypic correlations of 0.32 to 0.59 between linear measurements and adult body weight in Harnali sheep. Because the body weights of some animals were themselves estimated from body measurements, the present correlations are likely to be inflated and are not directly comparable with studies based entirely on directly measured weights.

Simple linear regression

Simple regressions of body weight on each body measurement are given in Table 3. All regression coefficients were highly significant (P < 0.01) at every age, and R 2 ranged from 0.868 to 0.980. Chest girth gave the highest R 2 at 0-3 (0.955), 7-12 (0.961) and above 36 months (0.980), height at withers at 4-6 months (0.935) and body length at 13-36 months (0.949). The best single predictor thus varied slightly with age, but chest girth was the most consistently useful. The regression coefficients increased with age, indicating that each centimetre of increase in a body dimension corresponded to a greater gain in body weight as the animals matured; this supports the use of age-specific rather than pooled equations.

Multiple regression

Multiple regressions of body weight on all three measurements are presented in Table 4. Fitting the three measurements together raised R 2 to 0.953-0.984, an improvement of only 0.004 to 0.018 over the best single predictor at each age. Such a small gain does not justify recording three measurements in the field instead of one. Height at withers, chest girth and body length were very highly intercorrelated (0.93-0.99), and variance inflation factors computed from the correlations in Table 2 exceeded 10 at every age, indicating severe multicollinearity. As a result, several partial regression coefficients were negative (e.g., -0.386 for chest girth at 4-6 months and -0.259 for height above 36 months). These negative coefficients are a statistical artefact of multicollinearity and do not imply that an increase in a body dimension reduces body weight. The multiple regression equations therefore have good overall fit but unstable individual coefficients and are not recommended for field use; stepwise or ridge regression could be used if more than one predictor is required.

Sex-wise regression and homogeneity of slopes

Sex-wise simple regressions (Table 5) showed close relationships between body weight and all three measurements in both sexes, with R 2 ranging from 0.825 to 0.989. Chest girth gave the highest R 2 in both sexes at 0-3, 7-12 and above 36 months, and in males at 13-36 months; the highest values were recorded above 36 months (0.989 in males and 0.977 in females). Sex-wise multiple regressions (Table 6) gave R 2 values of 0.938 to 0.992 and showed the same instability of partial coefficients as the pooled equations.

The test of homogeneity of slopes (Table 7) showed that the regression slopes of males and females did not differ significantly at 0-3, 4-6, 13-36 and above 36 months, so a single equation can be used for both sexes at these ages. At 7-12 months, the slopes differed significantly for all three measurements (P < 0.01), the female slopes being steeper than the male slopes. This age coincides with the sale of fastgrowing male lambs, and selective removal of males may have altered the body weight-body size relationship among the males that remained; this explanation, however, remains speculative. Separate equations for males and females are therefore recommended at 7-12 months.

Practical implications

Chest girth can be recorded quickly with a tape and, used in age-specific equations, gives a reliable estimate of body weight in local sheep. Such estimates can help farmers market animals on the basis of approximate live weight, allow field workers to monitor growth and nutritional status and to calculate doses of weight-dependent drugs, and permit selection for body weight in flocks without weighing facilities.

Limitations

The study was cross-sectional, so differences between age classes are comparisons among different animals rather than longitudinal records of the same animals. Body weights of some animals were estimated when direct weighing was not permitted, which is likely to have in-flated the correlations and R 2 values. The equations were not validated on an independent set of animals. They should therefore be validated on directly weighed animals from the same population before being adopted as a substitute for weighing.

Conclusion

Body weight of local (Dharuri) sheep of Beed district was closely and positively correlated with height at withers, chest girth and body length at all ages. Chest girth was the most consistent single predictor, giving the highest R 2 at three of the five ages, while adding height and body length improved prediction only marginally and produced unstable coefficients because of multicollinearity. A single equation served both sexes at all ages except 7-12 months. Age-specific chestgirth equations are recommended for estimating body weight of local sheep under field conditions, subject to validation on directly weighed animals.

Acknowledgement

AcknowledgementThe authors thank the sheep farmers of Adas, Wadhwani, Dharur and Neknoor for their cooperation during data collection.

References

1. Rani M, Ekambaram B, Punya Kumari B. Prediction of body weights of Nellore sheep under field condition. Indian Journal of Animal Research. 2016;50(1):133-136.
2. Kumar S, Dahiya SP, Malik ZS, Patil CS. Prediction of body weight from linear body measurements in sheep. Indian Journal of Animal Research. 2018;52(9):1263-1266.
3. Kumar A, Mallick PK, Misra SS, Sharma RC, Gowane GR. The morphometric scale to predict the live weight of Malpura sheep in semi-arid region of Rajasthan. Indian Journal of Animal Research. 2021;55(9):1105-1110.
4. Rather MA, Hamadani A, Shanaz S, Ahanger SA, Khalil M, Shah R. Prediction of bodyweight from linear body measurements in Gurez sheep. SKUAST Journal of Research. 2022;24(2):257-260.
5. Abbas A, Ullah MA, Waheed A. Body weight prediction using different data mining algorithms in Thalli sheep: A comparative study. Veterinary World. 2021;14(9):2332-2338.
6. Ibrahim A, Artama WT, Budisatria IGS, Yuniawan R, Atmoko BA. Regression model analysis for prediction of body weight from body measurements in female Batur sheep of Banjarnegara District, Indonesia. Biodiversitas. 2021;22(7):2723-2730.
7. Mekonnen T, Araya M, Tesfahun S. Technical options for body weight estimation: Prediction of body weights of indigenous sheep populations from their linear body measurements. Biotechnology Journal International. 2024;28(6):128-139.
8. Ravimurugan T, Thiruvenkadan AK, Sudhakar K, Panneerselvam S, Elango A. The estimation of body weight from body measurements in Kilakarsal sheep of Tamil Nadu, India. Iranian Journal of Applied Animal Science. 2013;3(2):357-360.
9. Firdaus F, Atmoko BA, Adinata Y, Panjaitan TS, Krishna NH, Widiyawati R, Makmur M. The meta-analysis of sheep body weight prediction with body measurement, breed and sex categories for practical livestock management purposes. Veterinary Integrative Sciences. 2025;23(3):e2025075.
10. Narsimham SAL, Narasimha Rao G, Gupta BR, Eswara Reddy C. Body weights and biometrical measurements of Deccani sheep. Indian Journal of Animal Sciences. 2003;73(8):930-933.
11. Steel RGD, Torrie JH. Principles and Procedures of Statistics: A Biometrical Approach. New York: McGraw-Hill; 1960. How to cite this article: Kukar K, Chauhan D, Shinde A, Deshmukh J. Relationship of body measurements with body weight at different ages in local (Dharuri) sheep under field conditions in Beed district of Maharashtra. International Journal of Advanced Biochemistry Research. 2026;10(10):23-28. doi: http s://www.doi.org/10.33545/26174693.2026.v10.i10a.10274