Livestock Research for Rural Development 25 (3) 2013 Guide for preparation of papers LRRD Newsletter

Citation of this paper

Predicting body weight from heart girth, height at withers and body condition score in Bos indicus cattle bulls of Uganda

C B Katongole, D Mpairwe, F B Bareeba, E Mukasa-Mugerwa* and C Ebong**

Department of Agricultural Production, Makerere University, P.O. Box 7062, Kampala, Uganda
* JNT/HPI Consultant P.O. Box 28491, Kampala, Uganda
** Rwanda Agriculture Board, P.O. Box 5016, Kigali, Rwanda


Heart girth (HG), height at withers (HW) and body condition score (BCS) were used to predict the live body weight (BW) of Bos indicus cattle bulls in Uganda using 24 Ankole and 24 Nganda growing bulls. The bulls were subjected to three feeding regimes for a period of 392 days (56 weeks). The data obtained on the three body measurements were fitted to simple and multiple linear regression equations to predict BW. Tests for homogeneity of slopes and intercepts showed no differences between the two breeds and across the three feeding regimes. Linear regressions indicated that all the three body measurements can be useful in predicting body weight; however, HG was the most accurate predictor as a single explanatory variable (R2 = 0.86) compared to HW (R2 = 0.76) and BCS (R2 = 0.55). The R2 value improved to 0.85 after HW and BCS were combined in the same equation. Quadratic and cubic effects of the three body measurements on BW were not significant. It is concluded that live body weight of Bos indicus cattle bulls can be accurately predicted from body measurements regardless of breed type and feeding regime.

Key words: Ankole bulls, indigenous breeds, Nganda bulls


The importance of indigenous cattle breeds (Bos indicus) to Uganda’s cattle stock cannot be overemphasized. According to MAAIF (1999), they account for over 96% of the total population. The major indigenous genetic resources are the Ankole Long-horn Sanga (which comprise 50%), the East African Short-horn Zebu (30%) and the Nganda cattle (16%), which originated from the natural cross-breeding of the East African Zebu with the Ankole. The Ankole (Photo 1) are characterized as large body frame cattle, while the East African Short-horn Zebu and the Nganda (Photo 2) are characterized as small and intermediate body frame cattle (Faulkner and Epstein 1957; Sacker and Trail 1966; Maule 1990). These breeds have acquired the adaptational characteristics that are essential for successful animal production in the tropics (Norval 1992; Ruane 1999). Consequently, they have an advantage over the introduced (exotic) breeds.

Photo 1. Ankole Long-horn bull

Photo 2. Nganda bull

The bulk of these breeds are kept by resource-poor farmers under traditional herding production systems. They are predominantly fed on natural pastures that are of a variable supply pattern and of poor quality with little or no supplementation. Additionally, although body weight is an important trait for several purposes including determination of live animal sale value, growth rate assessment, feed formulation, decisions regarding supplementary feeding, dosage of medications etc., it is not regularly monitored by most of the farmers rearing these cattle breeds. Weighing as a method of obtaining the body weight of animals is almost impossible given the absence of weighing devices among these farmers. Hence, the farmers use visual estimates, which is not always a reliable guide. Besides, these breeds are of different frame sizes, a parameter which has been reported to be highly correlated with growth rate, and hence live body weight (Dolezal et al 1993; Vargas et al 1999). It is for this reason that this study was conducted to derive predictive models of body weight from morphometric measurements in these cattle breeds. 

Materials and Methods

Location and climate of the study area 

The experiment was conducted at the National Crops Research Institute (NACRI), formerly known as Namulonge Agricultural and Animal Production Research Institute (NAARI). The institute is based at Namulonge in the central region of Uganda (Wakiso district). It is located 27 km north of Kampala at 0o31N and 32o35E and at an altitude of 1150 m above sea level. The mean annual rainfall is 1170 mm, which is bimodal in distribution, with April to May and October to December as the first and second rainy seasons, respectively. 

Experimental animals

A total of 48 growing bulls of two Bos indicus breeds (Ankole and Nganda) were used in the experiment. Each breed was represented by 24 bulls. The Ankole bulls (with an average initial body weight of 215.2±26.4 kg) were bought from a private farm in Mbarara district (western region of Uganda), while the Nganda bulls (with an average initial body weight of 117.8±32.5 kg) were selected from the cattle herd at NACRI. The bulls were weighed at the beginning of the experiment, and ear tagged with reference to the feeding regime to which each was assigned. 

Feeding regimes 

The Ankole (n = 24) and Nganda (n = 24) bulls were randomly assigned to three feeding regimes (Grazing, Grazing + SuppI and Grazing + SuppII) described in Table 1. The 24 bulls of each breed were divided into eight groups of three bulls according to body weight, and one bull from each group was randomly allocated to one of the three feeding regimes. This resulted into eight bulls per feeding regime per breed. 

Table 1. Description of the feeding regimes


Feeding regime



Grazing + SuppI

Grazing + SuppII

Basal diet#

Natural pastures

Natural pastures

Natural pastures

CP content, % DM




Supplement ingredients, % as mixed

Maize bran




Cottonseed cake




CP content, % DM




#All the 48 bulls were grazed on the natural pastures (mainly composed of Brachiaria and Panicum species mixed with Desmodium species) between 08:00 and 17:00 hours as one group

The supplements were offered in group pens daily, after grazing (overnight). The bulls offered the supplemented feeding regimes were grouped to balance for body weight so that bulls in the same pen were as similar in body weight as possible. Depending on the body weight change, the quantity of the supplements offered per bull per day was gradually increased from 1 to 2.5 kg in the respective group pens. 


After 392 days (56 weeks) of feeding, heart girth, height at withers and the corresponding body weights of all the bulls were measured before the animals were fed. The body weight was measured using a digital weighbridge. The heart girth was measured (using a tape measure) as the circumference immediately behind the front legs when the bulls were standing on the weighbridge. Height at withers was measured as the distance from the floor of the weighbridge to the withers. The bulls were also scored for body condition according to a 9-point scale for Bos indicus cattle (Nicholson and Butterworth 1986). 

Data analysis 

Linear regressions of body weight on heart girth, height at withers and body condition score were performed using the general linear model (GLM) procedure of SAS (2003). Quadratic and cubic effects of the independent variables on body weight were also assessed. Regression equations were established to predict body weight from the independent variables. The GLM procedure of SAS (2003) was used to compare regression coefficients between the two breeds and across the three feeding regimes.

Results and Discussion

The average body weight and measurements of the bulls after 392 days (56 weeks) of feeding are summarised in Table 2. The results are expressed as means ± standard deviation.   

Table 2. Average body weight and measurements (±sd) of Bos indicus cattle bulls of Uganda





Body weight, kg




Heart girth, cm




Height at withers, cm




Body condition score




Body weight was equally predicted regardless of breed and feeding regime. Tests for homogeneity of regression coefficients showed no differences between the two breeds and across the three feeding regimes (data not presented). Consequently, linear regression equations (simple and multiple) for predicting body weight were calculated for the pooled (overall) data (Table 3). The regression of body weight on heart girth produced the highest coefficient of determination (R2) value (0.86) followed by height at withers (0.76), and body condition score produced the lowest (0.55). Quadratic and cubic effects of the three body measurements on body weight were not significant. 

Table 3. Regression equations for predicting body weight from three body measurements in Bos indicus cattle bulls of Uganda



% change in R2

Simple linear regression



BW = -40.1 + 47.6BCS



BW = -561.0 + 6.8HW



BW = -401.3 + 4.7HG



Multiple linear regression



BW = -468.3 + 3.7HG+1.7HW



BW = -388.6 + 4.0HG + 14.2BCS



BW = -540.9 + 5.2HW + 26.0BCS



BW = -493.8 + 2.1HG + 2.8HW + 18.1BCS



BW, body weight; HG, heart girth; HW, height at withers; BCS, body condition score; *, P < 0.05; **, P < 0.01

When height at withers was added to heart girth in the prediction equation, the R2 value was raised by 1.2%, while the addition of body condition score raised the R2 value by 2.3%. Adding both height at withers and body condition score to the heart girth, raised the R2 value by 5.8%. Combining body condition score and height at withers in the same equation improved the accuracy of predicting body weight by 60% and 15.8% for body condition score and height at withers, respectively. 

Given that there was no bias observed in breed and feeding regime in the ability of the regression equations to predict body weight, it was not necessary to calculate separate linear regression equations for breed type and feeding regime. The relationship between body weight and the three body measurements was linear as the quadratic and cubic regression equations were found not to be statistically significant. 

Basing on the R2 values, regression equations containing either body condition score or height at withers as single variables did not account for enough of the variation in body weight to give an accurate prediction of body weight. The most effective prediction equation was that containing heart girth. This observation is in agreement with earlier studies (Wilson et al 1997; Alderson 1999; Goe et al  2001; Nsosoet al  2003; Mantysaari and Mantysaari 2008), which indicated that heart girth was the most reliable body measurement in the prediction of body weight for different types of animals. 

Prediction equations of body weight from multiple body measurements showed that the addition of a second or third explanatory variable to heart girth added little to the R2 value. This result indicates that in situations where body weight cannot be measured, heart girth alone can be effectively used to accurately estimate the body weight. Besides, when compared to height at withers, heart girth appears to be an easier measurement to make because it does not require the animal to adopt a position or posture with all its legs supported on the floor (Rodriguez et al 2007). Height at withers can be difficult to measure especially with nervous and temperamental animals. However, in a situation where body condition score and height at withers are the only available variables, they should be considered as two independent variables to accurately predict the body weight. 



The authors are grateful to the United States Agency for International Development (USAID), which funded this study through the Heifer Project International/Agriculture Business Systems (HPI/ABS). In the same breath, support from the Danish International Development Agency (DANIDA) through the Livestock Systems Research Programme (LSRP) is highly appreciated.


Alderson G L H 1999 The development of a system of linear measurements to provide an assessment of type and function of beef cattle. In: Galal S, Boyazoglu J (Eds.), Animal Genetic Resources Information Bulletin. Food and Agriculture Organization (FAO) of the United Nations and United Nations Environment Programme (UNEP). 25: 45 – 55. 

Dolezal H G, Tatum J D and Williams Jr F L 1993 Effects of feeder cattle frame size, muscle thickness, and age class on days fed, weight and carcass composition. Journal of Animal Science 71: 2975 – 2985. 

Faulkner D E and Epstein H 1957 The indigenous cattle of the British Dependent Territories in Africa, and of certain other African countries. Publication of the Colonial Advisory Council for Agriculture, Animal Health and Forestry. London, HMSO. 

Goe M R, Alldredge J R, Light and D 2001 Use of heart girth to predict body weight of working oxen in the Ethiopian highlands. Livestock Production Science 69: 187 – 195. 

MAAIF 1999 Meat Production Master Plan Study. Executive Summary. Republic of Uganda, Ministry of Agriculture, Animal Industry and Fisheries, Entebbe, Uganda. 

Mantysaari P and Mantysaari E A 2008 Relationship of body measurements and body condition score to body weight in modern Finnish Ayrshire cows. Acta Agriculturae Scandinavica, Section A - Animal Science 58(4): 170 – 178. 

Maule J P 1990 The cattle of the tropics. Centre for Tropical Veterinary Medicine, University of Edinburgh. Easter Bush, Roslin, Midlothian, Scotland. 

Nicholson M J and Butterworth M H 1986 A guide to condition scoring of Zebu cattle. ILCA, Addis Ababa. 

Norval RAI 1992 Host susceptibility to infection with Amblyomma hebraeum. Insect Science and its Application 13(4): 489 – 494. 

Nsoso S J, Aganga A A, Moganetsi B P and Tshwenyane S O 2003 Body weight, body condition score and heart girth in indigenous Tswana goats during the dry and wet seasons in southeast Botswana. Livestock Research for Rural Development 15(4)

Rodríguez C, Munoz L, Rojas H and Briones M 2007 New formula for bodyweight estimation of thoroughbred foals. Veterinary Record 161: 165 – 166. 

Ruane  J 1999 Selecting breeds for conservation. In: Oldenbroek JK  (Ed.), Gene banks and the conservation of Farm Animal Genetic Resource. DLO Institute for Animal Science, The Netherlands ISBN 90: 59 – 73. 

Sacker G D and Trail J C M 1966 A note on milk production of the Ankole cattle in Uganda. Tropical Agriculture TRIN 43: 247 – 250. 

Vargas C A, Olson T A, Chase C C, Hammond Jr A C and Elzo M A 1999 Influence of frame size and body condition score on performance of Brahman cattle. Journal of Animal Science 77: 3140 – 3149. 

Wilson L L, Egan C L and Terosky T L 1997 Body measurements and body weights of special-fed Holstein veal calves. Journal of Dairy Science 80: 3077 – 3082.

Received 12 February 2012; Accepted 8 February 2013; Published 1 March 2013

Go to top