Livestock Research for Rural Development 3 (1) 1991  Citation of this paper 
Prediction of sheep body weight in markets in the Far North Cameroon
E Thys* and J Hardouin**
* Directory of projects, studies and
training, Ministry of livestock, Fisheries and Animal Industries,
Yaoundé, Cameroon.
** Tropical Animal Production Service, Institute of Tropical
Medecine, Nationalestraat 155, B2000 Antwerpen 1, Belgium.
Summary
With the aim of providing an easy technique to predict the body weight of Poulfouli sheep in the market, regression equations have been calculated. The heart girth was retained as prediction variable. The formula of the allometric curve, explained 90.8% of the variation of the body weight in ewes and 86.5% in rams.
Resume
Prediction du poids vif de moutons presentes sur les marches de l'extremenord cameroun.
Dans le but de fournir une technique facile à appliquer et permettant de prédire le poids vif de moutons Poulfouli sur les marchés, des formules de régressions ont été testées. Le périmètre thoracique a été retenu comme variable prédictive ainsi que la formule de la courbe allométrique expliquant 90.8% de la variation du poids chez la brebis et 86.5% chez le bélier.
Resumen
Predicción del peso vivo de carneros presentados en los mercados del extremo norte del camerún.
Con objeto de establecer una técnica de aplicación fácil y que permita predecir el peso vivo de carneros Poulfouli en los mercados, se han ensayado formulas de regresión. El perímetro torácico ha sido usado como variable que permite predecir, como también la forma de la curva alométrica, explicando 90.8% de la variación del peso en la oveja, y 86.5% en el cordero.
KEY WORDS: Barymetry, liveweight, tropical sheep, measurements, Cameroon
Introduction
Weight determination is a major concern for the sheep buyers in the markets of the Far North Province of Cameroon, especially for rams selected for slaughter. The Animal Production Services are also interested in collecting regular weight data.
Due to the small size of the animals, weighing scales are indeed mostly recommended for this purpose. It is the most accurate method and the price of this type of scale is not too expensive. However, it appeared in the conditions of the Far North Cameroon that for longterm surveys on the market the breeders were rather opposed to this intervention.
The reasons seem to be both economical and psychological. Moreover, the major argument was that the discussions with the potential buyers would be affected by the determination of the exact weight. It was also felt that time was lost in the operation and the hanging of the sheep in the belt of the scale was considered to be dangerous for the animal. The pleasure of bargaining would also be reduced by weighing the animals.
Therefore, with the aim of avoiding these difficulties and the possible consequences on data collection, an attempt was made to find an easy and relatively accurate method to predict the weight of the sheep in the market. Barymetry, which consists in establishing a prediction formula based on linear body measurement, is very adapted to this aim.
It should also be kept in mind that the motivation of surveys is to determine the average weight of a group of animals more than the specific weight of individuals.
Materials and methods
Animals
The data were collected in the market of Maroua town (Diamaré Division). Only animals of the Poulfouli breed were considered. This thintailed sheep is related to the Djallonke or Dwarf African sheep from which it differs by its taller stature.
As sex status is an important factor of variation (Delage et al 1955) data on ewes and rams were collected separately. The ram sample included data measured during the rainy and the dry season; data on ewes were only collected during the dry season. An attempt was made to keep the composition of each sample close to the distribution of ages on the market (Table 1).
Measurements
Most of the studies made on cattle show that the heart girth is the most precise and easy to apply of the linear body measurements (Delage et al 1955; Vissac 1966; Dineur et al 1983). Orji and Steinbach (1981) came to the same conclusion in their study on the Nigerian Dwarf Sheep suggesting that determination of height of withers and body length do not improve significantly on formulae based on heart girth. This last measurement was consequently used in the study. The measurement was taken in the chest region as close to the forearms as possible using a pockettape.
Table 1: Distribution of the samples used for regression.  
Rams 
Ewes 

N 
percentage 
N 
percentage 

Sample size  167 
61.6 
70 
48.6 
Permanent incisors  
1 pair  70 
25.8 
26 
18.1 
2 pairs  26 
9.6 
30 
20.8 
3 pairs  7 
2.6 
15 
10.4 
4 pairs  1 
0.4 
3 
2.1 
TOTAL  271* 
100.0 
144 
100.0 
* 127 in rainy season, 144 in dry season
Results
Delage et al (1955) indicate for samples including different classes of age and growing stage, that curvilinear formulae give the best fit. The allometric curve (y = a*x^{b}) was used for this study and compared to the simple linear regression (y = a + bx).
Data were subjected to linear regression using a computer. In order to convert the curvilinear formula into a linear one, logarithmic transformation was applied (Sokal and Rohlf 1981). The standard deviation of the regression coefficient was computed and its signification tested by Ftest.
Table 2 shows the equations obtained for regression of body weight (Y) on hearth girth (X) for both sex groups.
Table 2: Regression equations of body weight (kg) on heart girth (cm) for rams and ewes in the market.  
Equations  Standard deviation b 
Fvalue 
Coefficient of Determination 
Rams:  
Y = 46.22 + 1.0383X  0.025 
1678* 
0.86 
lnY = 8.0007 + 2.6506(lnX)  0.0640 
1717* 
0.86 
Ewes:  
Y = 38.28 + 0.9268X  0.0278 
1111* 
0.88 
ln Y = 7.3839 + 2.5051(lnX)  0,0669 
1403* 
0.91 
* P<0.001
Discussion
Heart girth explained 86.5% of the variation of the weight of the rams and 90.8% of that of the weight of the ewes. The regression coefficients are all highly significant (P<0.001) and in both sex groups the coefficient of determination is improved by logarithmic transformation, although it is more marked for the ewes than for the rams. This can be explained by the distribution of ages in both sex groups, with the proportion of animals with no more than two permanent incisors significantly higher in rams than in ewes (chi^{2} = 25.63; P<0.001). The majority of the males are thus in the linear growing phase.
The reaction of the breeders to this method was positive and they accepted the small constraint. They compared the manipulation of the sheep with that of a tailor taking the waist measurement of a client.
Conclusion
Measuring heart girth with a pocket tape is easy, cheap, rapid, and well accepted by the breeders. It provides a prediction variable which explains 86.5% of the variation of body weight in rams and 90.8% in ewes. Based on the regression, conversion tables can be established and be made available to interested people (see table 3).
Table 3: Conversion of hearth girth (cm) into weight (kg) for rams and ewes using the logarithmic formula.  
Rams 
Ewes 

cm 
kg 
cm 
kg 
cm 
kg 
cm 
kg 

58 
15.8 
74 
30.2 
51 
11.8 
69 
25.1 

59 
16.6 
75 
31.3 
52 
12.4 
70 
26.0 

60 
17.3 
76 
32.4 
53 
13.0 
71 
27.0 

61 
18.1 
77 
33.5 
54 
13.6 
72 
27.9 

62 
18.9 
78 
34.7 
55 
14.2 
73 
28.9 

63 
19.7 
79 
35.9 
56 
14.9 
74 
29.9 

64 
20.6 
80 
37.1 
57 
15.6 
75 
30.9 

65 
21.4 
81 
38.4 
58 
16.3 
76 
32.0 

66 
22.3 
82 
39.6 
59 
17.0 
77 
33.0 

67 
23.2 
83 
40.9 
60 
17.7 
78 
34.1 

68 
24.1 
84 
42.3 
61 
18.4 
79 
35.2 

69 
25.1 
85 
43.6 
62 
19.2 
80 
36.4 

70 
26.1 
86 
45.0 
63 
20.0 
81 
37.5 

71 
27.1 
87 
46.4 
64 
20.8 
82 
38.7 

72 
28.1 
88 
47.8 
65 
21.6 
83 
39.9 

73 
29.1 
89 
49.3 
66 
22.5 
84 
41.1 

67 
23.3 
85 
42.3 

68 
24.2 
86 
43.6 

Acknowledgements
The authors are grateful to the veterinary agent Wouyie Dairou and the surveyor Bello who helped in the collection of the data.
References
Delage J, Poly J and Vissac B 1985 Annales de Zootechnie 3:219231
Dineur B, Thys E and Hardouin J 1983 Annales de Gembloux 89:914
Orji B I and Steinbach J 1981 Tropical Animal Health and Production 13:101106
Sokal R R and Rohlf F J 1981 Biometry, 2nd edition. W H Freeman and Company, New York.
Vissac B 1966 Annales de Zootechnie 15: 1545