Livestock Research for Rural Development 10 (3) 1998

Citation of this paper

Prediction of body weight on the basis of body measurements in mature indigenous chickens in Senegal

E F Guèye*, A Ndiaye** and R D S Branckaert***

Senegalese Institute of Agricultural Research (ISRA), PO Box 2057, Dakar-Hann, Senegal
*Present address (Animal Production Service, Animal Production and Health Division, FAO, Viale delle Terme di Caracalla, 00100 Rome, Italy)
** Directorate of Livestock (DIREL), Ministry of Livestock, PO Box 67, Dakar, Senegal
***Animal Production Service, Animal Production and Health Division, FAO, Rome, Italy

Abstract

Data on body weight and body measurements were individually collected from mature Senegalese indigenous fowls at two central poultry markets in Dakar, Senegal, between March and September 1998, with the objective to investigate the possibilities of using body measurements to predict the body weight of birds with high level of accuracy. Results, obtained from 502 males and 325 females, revealed that more males (60.7 %) were removed from fowl flocks for sale. The males showed higher body weights and body measurements (P<0.001).

Correlation coefficients between the body weight and the circumference of chest were strong and high (P<0.001 in males as well as in females). The body weight was also correlated to the body length (P<0.001 in males and P<0.01 in females). Thus, the circumference of chest and the body length are the body measurements that are most suitable for the prediction of the body weight.

Key words: Body measurement, body weight, indigenous chicken, Senegal

Introduction

Poultry keeping is of great importance to Senegalese households. Outside the urban centres, and especially in non-coastal areas, family poultry provides the population with a vital source of protein and income, and plays a key role within the context of many social events (eg: special banquets for family or for distinguished guests, cocks as alarm clocks for people, gifts, etc.) and/or religious ceremonies (eg: cocks as offerings to the deities). This industry is largely based on domestic chicken production. In 1995, the chicken population in Senegal was estimated to be 40 million (Anonymous 1996), and more than 70 % of it are constituted by the indigenous chickens (Guèye and Bessei 1997; Guèye 1998a). In Senegalese rural areas, indigenous chickens are generally raised in free-range and/or backyard systems, which are traditional extensive husbandry systems (Guèye 1997). Despite the importance of this subsistence poultry sector in Senegal, the marketing system is quite informal and poorly developed. The price of chicken largely depends on body weight. However, there is generally no weighing machine or scale available in poultry markets and in villages.

The objective of this study was to investigate the possibilities of using body measurements to predict the body weight of mature Senegalese indigenous chickens with high level of accuracy.

Materials and methods

Location

The data of the present study were collected from indigenous chickens at two central poultry markets of Dakar, the capital city of Senegal. These markets were chosen because high populations of indigenous chickens are available at these places at all times. Birds are brought there for sale regularly, especially on Mondays. Traders usually come to buy chickens in village markets and/or from village producers and/or from middlemen and resell them in markets of big cities, like those of Dakar.

Collection of data

Using a 20 kg weighing instrument and a measuring tape, information on body weight and body measurements was individually collected from all chickens found during 12 visits to the two markets from March to September 1998. Body measurements included body length (BL), circumference of chest (CC), femur (F), crus (C), tarsometatarsus (TM), as described by Salomon (1996). Measurements were carried out as follows:

Statistical analysis

As subclass frequencies were unequal, the General Linear Models Procedure was applied to the body weight and body measurements, and to calculate least square means. The following statistical model was used:

Yij = µ + Si + eij

where, Yij = estimated value for the body weight or body measurement; µ = population mean (mean that would exist if all classes had equal numbers); Si = fixed effect of sex; eij = residual error

Pearson’s correlation coefficients were estimated between BW and all body measurements. Data were subject to linear and multiple regression for predicting BW.

Results and discussion

Table 1: Mean values for body weight and body measurements

Mean ± standard deviation
(No. of birds)

Traits

(Males; n=502)

(Females; n=325)

Signif.
Body weight  (g)

1367 ± 304

1120 ± 239.9

***

Body measurements (cm)

Body length

40.8 ± 2.79

37.7 ± 2.30

***
Chest circumference

28.3 ± 2.81

25.7 ± 2.11

***
Femur

10.4 ± 1.57

9.5 ± 1.15

***
Crus

13.7 ± 1.60

12.0 ± 1.35

***
Tarsometatarsus

10.7 ± 1.09

9.20 ± 0.90

***
***   P<0.001

The mean values for BW and body measurements with respect to the sex of Senegalese indigenous chickens brought to markets for sale are presented in Table 1. In comparison with the females, more males (60.7 %) were sold from indigenous chicken flocks. This result confirmed previous findings reported by  Kuit et al (1986) in indigenous chickens and pigeons, Dafwang (1990) in indigenous chickens, ducks and guinea fowls and Guèye (1998a,b) in indigenous chickens.  Guèye (1998a,b).  However Otchere et al (1990) asserted that both males and females of indigenous chickens, guinea fowls, ducks and pigeons are sold. According to the report on pigeons by  Hassan and Adamu (1997), more males were also sold (54.9 % of 344 surveyed indigenous pigeons from 14 strains), although pigeon flocks are normally composed of pairs. Nevertheless, mating possibilities do exist since males and females are usually allowed to run together all the time in and around the compound of households, indigenous chicken flocks not being closed production units.

The males showed higher values of BW and body measurements (P<0.001). This agrees with reports of Ngou Ngoupayou (1990) in indigenous chickens, Missohou et al (1997) in indigenous chickens and Hassan and Adamu (1997) in indigenous pigeons.

Although information on the age of the birds could not be obtained because indigenous chicken traders were ignorant of it, even most indigenous chicken keepers do not know the exact age of their birds (Buldgen et al 1992), the average body weights of the surveyed indigenous chickens indicated that they were probably about 6 months of age, because Sall (1990) reported similar values for indigenous chickens at this age under extensive husbandry systems of villages.

Throughout the year, indigenous chickens are regularly sold at markets. During the period of this study, prices of surveyed indigenous chickens were not affected by the sex of birds. Using a dressing percentage of 80 %, dressed birds at markets were estimated. Their prices ranged from US$ 2.25 to 3.30 per kg. They are consistent with the estimated average indigenous chicken meat prices in 1995 in Dakar which varied from US$ 2.54 to US$ 3.93 per kg at markets and supermarkets, respectively (Guèye 1998a). According to Guèye (1998a), these amounted to premiums of about 13 % at markets and 27 % at supermarkets in comparison with the estimated prices of meat from commercial chickens. Thus, well-off people who shop in supermarkets appear willing to pay more in order to get indigenous chicken meat (Guèye 1998a). Furthermore, compared with chicken meat from commercial flocks derived from imported stocks, those from indigenous stocks are strongly preferred by Senegalese consumers (Guèye and Bessei 1996, 1997).

Table 2: Correlation coefficients between body weight (BW) and body measurements (values for males are above the dotted line and those for females are below)
BW BL CC F C TM
BW - 0.68*** 0.74*** 0.43*** 0.34*** 0.52***
BL 0.37** - 0.48*** 0.66*** 0.72*** 0.68***
CC 0.80*** 0.27* - 0.24* 0.18ns 0.26**
F 0.00ns 0.46*** –0.14ns - 0.68*** 0.62***
C 0.10ns 0.59*** 0.00ns 0.70*** - 0.73***
TM 0.21ns 0.54*** 0.01ns 0.48*** 0.61*** -
ns = non significant (P>0.05); * = P<0.05; ** = P<0.01; *** = P<0.001

The correlation coefficients between BW and body measurements for both males and females are given in Table 2. All significant values of correlation coefficients were found to be positive. In males, all body measurements were strongly correlated to the BW of mature Senegalese indigenous chickens (P<0.001), while the BW of females was correlated only to BL (P<0.01) and to CC (P<0.001). With regard to the BW of birds, highest correlation coefficients were found with BL and CC, in males as well as in females. In indigenous pigeons, Hassan and Adamu (1997) also observed that body length as well as chest width were strongly and significantly correlated to body weight.

Table 3: Prediction of body weight (BW, g) on the basis of body length (BL, cm) or/and circumference of chest (CC, cm)
Sex Regression equations R2 Significance
Male BW = –1653 + 74.0 BL 0.46 ***
Male BW = –899 + 80.2 CC 0.55 ***
Male BW = –2138 + 45.4 BL + 58.4 CC 0.69 ***
Female BW = –346 + 38.8 BL 0.14 **
Female BW = –1230 + 91.3 CC 0.64 ***
Female BW = –1747 + 17.2 BL + 86.1 CC 0.67 ***
** = P<0.01; *** = P<0.001

The prediction equations consisting of body measurement variables are presented in Table 3. The comparison of R2 values for such regression equations indicated that, when multiple regression techniques based on BL and CC were used instead of linear regressions, there were improvements of about 50 % to 25 % in R2 values for males, and about 379 % to 5 % for females. Further addition of body measurements in prediction equations did not make any substantial improvement in R2 values. BL and CC represent higher values in comparison with the other body measurements and are easier to measure. Thus relative errors, especially due to the "measurer", are lower. It is therefore recommended to use multiple regression techniques based on BL and CC.

Conclusion

This study was undertaken with the aim of using body measurements to predict the body weight of mature Senegalese indigenous chickens. Thus, only using a measuring tape, it is possible to predict the BW of these birds with high level of accuracy, BL and CC being body measurements that are most suitable for this purpose. The measuring tape is much more available than a weighing machine or a spring balance in poultry markets. These prediction formulas can help consumers, who generally are not indigenous chicken keepers, to have an accurate, easy, cheap and rapid estimation of the body weight of the birds. Furthermore, in spite of the shortcomings of the source of data, an attempt was made to describe the Senegalese indigenous chicken which shows a considerable variation in phenotypic traits that are obviously affected by genetic components, in addition to the major impact of environmental factors.

References

Anonymous  1996 FAO Production Yearbook Vol. 49 - 1995. FAO Statistics Series No. 130, FAO, Rome, Italy

Buldgen A, Detimmerman F, Sall B and Compère R 1992 Study of demographical and zootechnical parameters of local hens in the Ground-Nut Basin of Senegal. Revue Élev. Méd. vét. Pays trop. 45 (3-4): 341-347 (in French)

Dafwang I I 1990 A survey of rural poultry production in the Middle Belt region of Nigeria. In: Proceedings of an International Workshop on Rural Poultry Development in Africa (Ed. E B Sonaiya), Ile-Ife, Nigeria, pp 221-235

Guèye E F  1997 Diseases in village chickens: Control through ethno-veterinary medicine. ILEIA Newsletter 13 (2): 20-21

Guèye E F  1998a Village egg and chicken meat production in Africa. World’s Poultry Science Journal 54 (1): 73-86.

Guèye E F  1998b Poultry plays an important role in African village life. World Poultry 14 (10): 14-17

Guèye E F and Bessei W  1996 Gefluegelhaltung in Afrika: Bedeutung und Perspektiven. DGS Magazin, Woche 31: 38-40

Guèye E F and Bessei W  1997 The importance of poultry farming in Senegal. Animal Research and Development 45: 82-88.

Hassan W A and Adamu U A  1997 Pigeon genetic resources in semi-arid zone of Nigeria: Initial results from characterization studies. Paper submitted at INFPD Workshop and General Meeting, 9-13 December 1997, M’Bour, Senegal

Kuit H G, Traore A and Wilson R T 1986 Livestock production in Central Mali: Ownership, management and productivity of poultry in the traditional sector. Tropical Animal Health and Production 18: 222-231

Missohou A, Sow R S and Ngwe-Assoumou C  1997 Caractéristiques morphobiométriques de la poule du Sénégal. Paper presented at INFPD Workshop and General Meeting, 9-13 December 1997, M’Bour, Senegal

Ngou Ngoupayou J D 1990 Country report on small-holder rural poultry production in Cameroon. In: CTA Seminar Proceedings, Volume 2, Smallholder Rural Poultry Production, Thessaloniki, Greece, pp 39-47

Otchere E O, Adeoye A T, Gefu J O and Adewuyi A A 1990 Preliminary observations on village poultry production, North-Central Nigeria. In: Proceedings of an International Workshop on Rural Poultry Development in Africa (Ed. E B Sonaiya), Ile-Ife, Nigeria, pp 196-200

Sall B  1990 Contribution à l’étude des possibilités d’amélioration de la production en aviculture traditionnelle: mesure du potentiel de la race locale et des produits d’un croisement améliorateur. Mémoire d’Ingénieur Agronome, INDR, Thiès, Sénégal

Salomon F -V  1996 Allgemeines Bauprinzip und aeussere Anatomie der Voegel. In: Lehrbuch der Gefluegelanatomie (Hrsg. F. -V. Salomon). Gustav Fischer Verla, Jena, Germany, pp. 19-25

Received 3 December 1998

Go to top