Genetic evaluation of growth traits in crosses between two ecotypes of Nigerian local chicken

O M Momoh and C C Nwosu*

Department of Animal Breeding and Physiology, University of Agriculture, P.M.B 2373, Makurdi. Benue State, Nigeria
* Department of Animal Science, University of Nigeria, Nsukka. Enugu State, Nigeria
momohmichael@yahoo.com

Abstract

Genetic evaluation of the Nigerian local heavy chicken ecotype (HE) and its F₁ crosses with the light ecotype (LE); HE X LE as the main cross (MCX) and LE X HE as the reciprocal cross (RCX) was carried out at the Poultry Farm of the Department of Animal Science, University of Nigeria, Nsukka. The objective of the study was to provide estimates of heritability of some growth traits and the genetic correlations between them in HE and its crosses under improved management. A total of 214 pedigree hatched day old chicks of HE, 190 day old chicks of MCX and 185 day old chicks of RCX from a total of 15 sires were used in the study. Body weight (BW), average daily weight gain (DG) and feed conversion ratio (FCR) were measured at 4-weekly intervals in all the genetic groups. Heritability estimates and genetic correlation between the traits were obtained.

Body weight in HE, MCX and RCX were on the average lowly to moderately heritable. Heritability estimates of daily gain were generally low in HE and RCX with a range of h² = 0.04 ± 0.13 to 0.12 ± 0.14. feed conversion ratio had moderate h² estimate in all the genetic groups. In all the groups, genetic correlation estimates between body weight and body weight gain were high, positive and highly significant (P < 0.001).

The low to moderate heritability of growth traits in the Nigerian local heavy chicken ecotype and its crosses as well as the variations in additive genetic effect reflect genetically unimproved populations.

Key words: Genetic correlations, heritability, main crossbred, reciprocal crossbred

Introduction

The local chickens of Nigeria numbers about 103 million (RIM 1992), 85% of which are found in the north and the rest being in the southern part of the country. They exhibit large variations in body shape and sizes and as noted by Oluyemi et al (1982) fowls from the southwest region may be different from those in other parts of the country. The local chickens of Nigeria for convenience can, therefore, be classified into light and heavy ecotypes on the basis of body weight and body size. The light ecotype represents the chicken type from the Swamp, Rainforest and Derived savannah agro-ecological zones whose mature body weight ranges from 0.68 to 1.5 kg and the heavy ecotype are those of the Guinea savannah, Sahel savannah and some Motane regions whose mature body weight ranges from 0.9 to 2.5 kg (Atteh 1990). These ecotypes have lived and produced for several years in the Nigerian environment under natural selection, therefore it is expected that they are adapted to their environment and may carry some genes favourable to the poultry industry in the future. However, little is known about their genetic properties, particularly the heavy ecotype.

In order to establish breeding programme it is necessary to count with heritability estimates of traits of economic importance and genetic association between them. This is because the degree of heritability allows one to estimate the amount of improvement by selection and genetic association can dictate method of selection.

Asuquo and Nwosu (1987), Ndofor (2003) and Udeh and Omeje (2004) provided heritability estimates of some growth traits of the Nigerian chicken type described here as light ecotype. No reported values of genetic parameter estimates are available for the chicken type described as heavy ecotype. The goal of this study is to estimate heritability of growth traits and genetic correlations among them in the Nigerian heavy chicken ecotype and its crosses with the light ecotype.

Materials and methods 

This study was carried out at the local chicken unit of the poultry farm of the Department of Animal Science, University of Nigeria, Nsukka. Nsukka is located on latitude 05⁰22’ North and longitude 07⁰24’ East with annual rainfall ranging from 986 to 2098mm (Inyang 1978). The natural day length for Nsukka is 12-13 hours and average annual maximum and minimum temperatures are 29.7^oc and 21.0^oc, respectively. The relative humidity ranges from 34 to 78% (Monanu 1975).

The foundation stock for the study consisted of the heavy chicken ecotype (HE) gathered from the rural areas of Obudu, a Montane region in Cross River State, south east Nigeria and the Guinea savannah agro-ecological zone of Northern Nigeria and the light ecotype (LE) obtained from the Swamp, Rainforest and Derived savannah of southern Nigeria. The HE and LE were being maintained on the farm as two separate non-pedigreed, unselected and unimproved random mating populations.

Generation of experimental birds

Fifty (50) HE hens randomly selected from the foundation stock were replicated into 5 deep litter pens with each pen containing 10 hens and a wing banded HE cock. Another set of randomly selected 50 HE hens were replicated into 5 deep litter pens with each pen containing 10 hens and a wing banded LE cock. Similarly, 50 randomly selected LE hens were replicated into 5 pens with each pen containing 10 hens and a wing banded. HE cock. The mating arrangement and the genetic groups produced are as set out below.

The parent (breeding) stock was fed with a formulated breeders diet containing 16% crude protein and 2600 ME kcal/kg of feed at the rate of 70g per bird per day. Water was provided ad lib.

Hatching eggs were collected and pedigree hatched by a process of backward integration in local chicken natural incubation using the basket system as described by Momoh et al (2004). The natural incubation was used to hatch the eggs because of absence of a functional artificial incubator.

Three batches of hatching of day old chicks were carried out each for all the genetic groups at two weekly intervals. These gave the total of 214 day old chicks for the straight breed heavy, 190 day old chicks for main crossbred and 185 day old chicks for the reciprocal crossbred

Management of experimental birds

On hatching, chicks were wing banded and weighed. Each hatch in each genetic group was reared separately from day old to 20 weeks on deep litter. The chicks were brooded for 8 weeks (0-8 weeks) on a formulated chick mash (20% crude protein and 2800 ME kcal/kg of feed) and reared from the 9^th to 20^th week on a formulated grower’s diet (16% crude protein and 2679 ME kcal/kg of feed). During brooding and rearing periods, all chicks were fed and provided with water ad libitum. In addition, they were medicated similarly and regularly and subjected to the same standard managerial, hygienic and climatic conditions.

Traits evaluated

Body weight, body weight gain and feed conversion ratio were evaluated for the genetic groups. Individual body weight was taken at hatch (BW 0), 4 weeks (BW 4), 8 weeks (BW 8), 12 weeks (BW 12), 16 weeks (BW 16) and 20 weeks of age (BW 20). Body weight gain was taken at 4 weekly intervals and expressed as daily weight gain at 0-4 weeks (DG 4), 4-8 weeks (DG 8), 8-12 weeks (DG 12), 12-16 weeks (DG 16) and 16-20 weeks (DG 20). Feed conversion ratio expressed as the ratio of mean feed consumed to mean weight gained was also calculated at 4 weekly intervals and designated as, 0 – 4 weeks (FCR 4), 4-8 weeks (FCR 8), 8-12 weeks (FCR 12), 12-16 weeks (FCR 16) and 16-20 weeks (FCR 20).

Analytical procedure

The data were subjected to heritability estimates and genetic correlation analysis using the Mixed Model Least squares and Maximum Likelihood computer programme of Harvey (1990). The reduced sire model (Becker 1992) was used to fit the data.

Ү _ij = μ + a_i + е _ij

Where,

Ү _ij =   Observation on the j^th bird of the i^th sire.

μ  =     Overall population mean.

a_i =      Random effect of the i^th sire (I = 1…5)

e_ij =      Residual random error.

The Harvey programme computes estimates of heritability and genetic correlations as:

Heritability:

                                          and                  

Genetic correlation:

Where:

h²   =Heritability estimate

γG = Genetic correlation coefficient

σ_s² = Sire variance component

σ²_e = Error variance component

Сov_s(x,y) = sire covariance component between traits x and y.

Results 

Heritability estimates of body weight at various ages derived from sire variance components for the heavy ecotype and the F₁ crossbreds are presented in Table 1.

Generally, there were age to age variations in heritability estimates of body weight in all the genetic groups with the two crossbred groups showing less variability when compared with the ‘straight bred’ heavy ecotype.

The heritability estimate of the heavy ecotype increased from 0.18 ± 0.19 at 4 weeks to 0.43 ± 0.26 at 8 weeks and thereafter declined to 0.16 ± 0.18 at the 16^th week and a sharp rise to 0.30 ± 0.23 at the 20^th week. Similarly, heritability estimate of the reciprocal crossbred increased gradually up to 12 weeks and thereafter declined gradually till the 20^th week. On the average, the body weight of the heavy ecotype and the crosses could be described as being lowly to moderately heritable.

The values of heritability estimates of body weight gain in the heavy ranged from h² = 0.04 ± 0.13 at 0 - 4 weeks to 0.12 ± 0.14 at the 16 – 20^th week. The estimates of the main crossbred were high in all the ages. The estimates for the reciprocal crossbred increased gradually with age. Apart form the main cross, the heritability estimates of body weight gain were generally low.

In all the genetic groups, there was increasing trend in the value of estimates of heritability of feed conversion ratio with age. Both the heavy and the main crossbred increased gradually from 0 - 4 weeks to a peak at 12-16 weeks and thereafter declined. On the other hand, the reciprocal crossbred increased in heritability estimates from 4 - 8 weeks to 16 - 20 weeks of age. On the average, the heritability estimates of feed conversion ratio were moderate in all genetic groups with large standard errors of estimate.

Table 2 presents the genetic correlation estimates between body weight, body weight gain and feed conversion ratio of heavy ecotype and its F₁ crosses at specific ages.

In all the groups, genetic correlation estimates between body weight and body weight gain were high, positive and highly significant (p < 0.001), although most the values in the reciprocal crosses were outside the parametric range of 1 or 100%. Conversely, genetic correlation between body weight and feed conversion ratio and between body weight gain and feed conversion ratio were negative (p < 0.001) in all the genetic groups. The main crossbred groups exhibited a more consistent and less variable (between ages) estimates of genetic correlations.

Discussion 

Heritability estimate of 20 week body weight obtained for the heavy ecotype is in close agreement with the value (h²= 0.33 ± 0.34) reported by Udeh and Omeje (2004) but slightly lower than the value of 0.46 ± 1.21 reported by Ndofor (2003) for the Nigerian local chicken. The heritability estimates of 4 -, 8-, and 12- week body weight of the heavy and its crosses in this study generally fall within the ranges reported by Kinney (1969) who using data from the literature and estimates based on ANOVA procedures reported mean heritability values of 0.43 (range 0.19 – 0.66), 0.38 (range 0.01 – 088) and 0.40 (range 0.38 – 0.73) for body weight at 4, 8 and 12 weeks old chickens, respectively.

Judging from the values of heritability estimates, it can be inferred that there is less genetic variability in the crossbreds than there is in the heavy ecotype. The heavy ecotype, therefore, offers more chances for selection than the crossbred groups. Moreover, crossbred groups generally may not be involved in direct selection because of problems of heterozygosity. The age to age variations in estimates of body weight heritability in all the groups which showed increasing trend with age to a maximum and thereafter declined are indicative of changes in the contributions of environmental deviation components of total variation in body weight. Chambers (1990) had noted that heritabilities for body weight of broilers tended to increase with age. Greater environmental influence (particularly those resulting from maternal influences) on body weight is expected at earlier ages.

The heritability values of 0.43 and 0.30 obtained for the heavy ecotype at 8 and 20 weeks respectively indicate that the heavy chicken ecotype has dual penitential to be selected as meat type or egg-type bird because 8 weeks and 20 weeks of age are broiler and layer ages, respectively. These moderate to high heritabilities indicate that response to selection at the 8^th or 20^th week could be rapid.

The heritability estimates of body weight gain and feed conversion ratio obtained in this study for all the genetic groups are far below the value (0.76 for weight gain and 0.58 for feed conversion ration) reported by Ikeobi and Peters (1996) for improved meat-type chickens in Nigeria. This is probably due to the fact that the local chickens are largely uncharacterized and unimproved as neither meat nor egg-type bird. Except in the main cross where heritabilities of body weight gain appeared high, the results of body weight gain in the other groups indicate that selection for body weight gain in the local chicken or its crosses might not result in any appreciable improvement unless some breeding work is carried out. Crossbreeding and family selection may be more appropriate in improving weight gain.

The significant (p<0.001) positive genetic correlations observed between body weight gain in all the genetic groups are in agreement with the findings of Ikeobi and Peters (1996) for these traits in meat-type chicken. However the negative genetic relationship between body weight and feed conversion ratio and between body weight gain and feed conversion ratio in this study contrasts with the result of Ikeobi and Peters (1996) who reported positive genetic correlation between these traits in meat-type chicken. The negative relationship may have arisen due to the fact that these traits have not been improved by way of selection in the local chicken ecotypes. Moreover, the negative genetic relationship may be beneficial in the local chicken since increases in feed conversion ratio would not be desirable. This is because increases in feed conversion ratio would entail the birds consuming more feed per unit gain in body weight.

Conclusions

• Body weight in the heavy chicken ecotype of Nigeria and its crosses are lowly to moderately heritable. Heritability estimates of body weight in the heavy ecotype indicate that selection for body weight at 8 weeks of age is expected to result in rapid improvement in the 8 weeks body weight, which is regarded as broiler age.

  
  

• Heritability estimate of body weight gain are generally low in the heavy ecotype and reciprocal cross while feed conversion ratio is moderately heritable in all the genetic groups. The low to moderate heritability of growth traits in the Nigerian heavy chicken ecotype and its crosses as well as the wide variation in additive genetic effects reflect genetically unimproved populations. This also means that environmental factors are more important in influencing growth traits in these populations than genetic factors.

Acknowledgement

The authors are grateful to the Department of Animal Science, University of Nigeria, Nsukka for the kind permission to use the facilities at the department for this study.

References 

Asuquo B O and Nwosu C C 1987 Heritability and correlation estimates of body weight in the local chicken and their crosses. East African Agricultural and Forestry Journal 52 (4) 267-271

Atteh  J O 1990  Rural poultry production in western middle-belt region of Nigeria. In: Rural Poultry Production in Africa. Editor: Sonaiya E B, Proceeding of an international workshop on rural poultry in Africa. Ile-Ife Nigeria 13-16 November 1989. 211-217

Becker W A 1992 Manual of Quantitative Genetics. 5^th Edition USA Academic Enterprise Pullman, 189pp.

Chambers J R 1990 Genetics of growth and meat production in chicken. In: Poultry Breeding and Genetics. Editor: Crawford R D, Amsterdam, The Netherlands, Elsevier 612-633

Harvey W R 1990 Mixed Model Least-squres and Maximum Likelihood Computer programme. Ohio State University Columbus (Mimeo)

.

Ikeobi  C O N and Peters S O 1996 Strain differences in genetic parameter estimates for growth traits in meat-type chicken. Nigeria Journal Animal Production 23 (2): 103-106

Inyang P.E.B 1978 The Climate of Nsukka Environs. In: The Nsukka Environment.(Editor: Ofomata, C.E.K)  Enugu Fourth dimension publishers. 152pp

Kinney T B Jr. 1969 A Summary of Reported Estimates of  Heritabilities and Genetic and Phenotypic Correlations for Traits of Chicken. Agricultural Handbook No.363. Washington  D.C. USDA Agricultural Research Services. 44pp.

Momoh O M, Tule J J and Nwosu C C 2004 Backward integration in natural incubation of local chickens using the basket system. Proceedings of the 29^th annual conference of the Nigerian society for Animal production. 21^{st _} 25^th March 2004. Usman Danfodio University Sokoto 29: 36-39

Monanu P.C 1975 Temperature and Sunrise. In: Nigeria in Maps: Eastern State. Editor: Ofomata, C  .K.  Benin city Ethiope publishing house. 146pp

Ndofor H M 2003 Estimates of genetic parameters of growth traits of local chicken ecotypes reared in Nsukka. M.Sc. thesis. Department of Animal science, University of Nigeria, Nsukka. 72pp

Oluyemi J A, Longe G O and Songu T 1982 Requirement of the Nigerian indigenous fowl for protein and amino acids. Ife Journal of Agriculture 4: 105-110

RIM 1992 Nigerian Livestock Resources. Volume II. National synthesis Annex publication. Resources Inventory Management Ltd.

Udeh I and Omeje S S I  2004  Heritabilities and genetic correlations between body weight and pecking behaviour in Nigerian native chickens. Proceedings of the 9^th Annual conference of Animal Science Association of Nigeria (ASAN) Abakiliki 13-16 September 265 - 266

Received 20 March 2008; Accepted 2 July 2008; Published 3 October 2008

Go to top