Livestock Research for Rural Development 15 (1) 2003

Citation of this paper

Genetic parameters for body weights of Creole chickens from Southeastern Mexico using an animal model

Prado-González E A, Ramírez-Avila L and Segura-Correa J C

Facultad de Medicina Veterinaria y Zootecnia, Universidad Autónoma de Yucatán,
Km 15.5 carretera Mérida-Xmatkuil. Mérida, Yucatán, México.


Genetic parameters for body weights up to 16 weeks of age from a Creole chicken population in Yucatan, Mexico were estimated. Information in body weights was obtained from 435 chicks, progeny of 34 roosters and 108 scavenge hens bought from different rural communities of Yucatan. Chicks were identified individually at birth, using numbered wing-bands, and weighed. Parents and chicks were fed commercial diets with 16.5 to 21.0% crude protein. Data on birth weight and weights at 4, 8, 12 and 16 weeks of age were analyzed by DFREML procedures using two animal models: Animal model for birth weight and 4 week body weight included the additive direct effect of animal, the additive maternal effect, the covariance of additive direct-maternal effects and the common environmental effect of the hen. Model for body weights at 8, 12 and 16 weeks of age was similar to the former model, except that the covariance between the direct and maternal additive effects was not included. Estimates of direct heritability were low (0.07 to 0.21), the lowest and highest corresponding to 16 and 8 weeks old birds, respectively. Maternal effects were important at birth and 4 weeks of age. A negative genetic correlation between additive direct-maternal effects was found at birth (-0.15). This study showed that additive heritabilities are low for body weight of Creole chickens during rearing. Also showed that direct maternal and common environmental are not important sources of variation after 4 weeks of age. 

Keywords: animal model, Creole chicks, heritability, Mexico, variance components. 


The Creole chickens in Mexico have been under natural selection for almost five centuries, therefore it is expected that they are adapted to its environment and may carry some genes favorable to the poultry industry in the future. However, little is known about its performance and less about the genetic variability and genetic parameters for this biotype. In order to establish breeding programs it is necessary to count with heritability estimates for the traits to be improved; because the degree of heritability allows us estimate the amount of improvement by selection. Therefore the higher the heritability of the trait the higher the response to selection. Heritabilities have been estimated for a number of chicken traits. However, it is necessary to remember that the heritability is a property not only of the trait but also of the population and the environmental conditions surrounding the animals (Falconer 1981). 

Heritability estimates reported in the literature for body weight in commercial chickens vary considerably (Kinney 1969). Except for the heritability of birth weight reported in Creole chickens by Hernandez and Segura (1994), there is not much information on heritability of body weights in Creole chickens in Mexico. ANOVA procedures are appropriate to estimate heritability provided the data set is large and there is good balance or structure of the data. However, animal model procedure is more appropriate when maternal and permanent environmental effects are also to be estimated, parents are related and selection is going on in the population under study. Reliable estimates of genetic parameters are needed to accurately predict the direct or correlated response to selection. 

The objective of this study was to obtain estimates of additive direct genetic, additive maternal and environmental maternal effects for body weights at birth, 4, 8, 12 and 16 weeks of age in Creole chickens in Yucatan, Mexico. 

Material and methods 

The study was carried out at the Faculty of Veterinary Medicine and Animal Science of the University of Yucatan, in Merida, Mexico. The region has a humid tropical climate, Aw1 (Garcia 1988). Average annual temperature is 26°C with a range from 18°C in December to 32 °C in May. The population under study arose from the insemination of 140 scavenging Creole hens with semen from 35 Creole roosters bought from different communities of Yucatan, Mexico. Insemination was done twice a week and each rooster was mated to the same four hens. Adult birds were allocated in individual cages in an open house and fed a commercial diet (16.5% CP, 2900 Kcal ME/kg). Chicks were obtained from four hatches, one or two weeks apart from June to July of 2001. At hatch, pedigreed chicks were wing-banded and housed on deep litter in an open house from 1 to 112 days of age at a housing density of 10 birds/m2. The hatching chicks were placed in a confined area of artificial heat for four weeks (temperature 32° C). At the age of 29 days birds were allowed the whole pen space (stocking density of 8 birds/m2). Mixed chicks were fed ad libitum on a starter diet (containing 21.0% CP and 3000 kcal ME/kg) from hatching to 4 weeks of age, followed by a grower diet (18% CP and 2900 kcal ME/kg) to 12 weeks of age and a diet with 16% CP and 2850 kcal ME/kg during the final growth phase. Water and feed were available ad libitum to the birds and they were reared under decreasing natural light conditions (13 to 11 hours of natural light).  

Body weights were recorded at day of hatching (Day 0) and at 4, 8 12 and 16 weeks of age. Birds were weighed individually on an electronic balance, within 0.1 g precision. Chicks were classified by sex (male, female) and by genotype, according to the presence of absence of the Na (naked neck) gene (naked neck homozygous NaNa, naked neck heterozygous, Nana and normal feathering, nana). Sex was determined at week 10 by phenotypic appearance. Some dams either did not lay or had no chicks at hatching and records from chicks that lost their wing-bands before sexing were not included. After data editing, a total of 435 chicks of 34 sires and 108 dams were available for analysis.  

Statistical analysis  

Variance components and genetic parameters were estimated for each body weight data set using DFREML program (Meyer 1998). The animal model for birth weight and 4 weeks body weight included the fixed effects of hatch number, sex and genotype of the chicken and the random effects of animal additive direct, additive maternal and common environmental effect of each dam common to their all progeny, and residual environmental effect. 

The animal model in matrix notation was: 

y = Xb + Z1a + Z2m + Z3c + e 

where:  y is a vector N x 1 of observations for birth weight,
X is the design matrix of 0’s and 1’s describing which elements of b correspond to observations in y;
b is an unknown vector of fixed effects to be estimated;
Z1 is the design matrix of 0’s and 1’s relating elements of a to observations in y;
is the vector of random additive direct genetic effects;
Z2 is the design matrix of 0’s and 1’s that associates m to observations in y;
m is the vector of random genetic maternal effects;
Z3 is the design matrix of 0’s and 1’s that relates elements of c to observations in y;
c is the vector of random maternal common environmental effects; and
e denotes the vector of random residual effects (temporal environmental).

Additive direct and maternal effects were assumed normally distributed with mean 0 and variance A and A, respectively, where A is the numerator relationship matrix and  and are the additive direct and additive maternal variances, respectively. Common environmental effects of the dam and residual were assumed to be normally distributed with mean 0 and variances Id and In, respectively, where Id and In are identity matrices with orders equal to the number of hens and chick records, respectively  and  and are maternal common environmental and residual variances, respectively. Estimates of additive direct (h2), additive maternal (m2) and common environmental (c2) heritabilities were calculated as ratios of estimates additive direct (),  additive maternal () and common environmental maternal () variances, respectively to the phenotypic variance (). The direct-maternal correlation (ram) was computed as the ratio of the estimates of direct-maternal covariance () to the product of the square roots of estimates of  and .

The animal model for body weight at 8, 12 or 16 weeks of age was similar to the model for birth weight except that the correlation between additive direct and additive maternal effects was not considered, because in previous run was not estimable. 

Results and discussion 

The results of sex and genotype of the growth of the Creole chickens will be published somewhere else (Segura-Correa and Juarez-Caratachea 2002 in press). Briefly, there were differences between sexes, but not among hatches and genotypes. The general least square means were 1484 and 1936 g for females and males, at 16 weeks, respectively. For genotypes the means were: 1721, 1657 and 1750 g for the homozygotes naked neck, heterozygotes naked neck and normal feathered birds. 

Genetic parameter estimates for body weights of Creole chickens are presented in Table 1. Additive direct heritabilities were low to moderate (range 0.07 to 0.21). The lowest and highest heritabilities were for the 16 and 8 weeks old chickens, respectively. Low heritabilities means that dominance, epistatic and environmental effects are more important than genetic additive effects on body weight of Creole chicks, at least under the present conditions of this study. Kinney (1969) reported, using data from the literature and estimates based on ANOVA procedures, mean heritability values of 0.43 (range 0.19-0.66), 0.38 (range 0.01-0.88), 0.40 (range 0.38-0.73) for body weights of 4, 8 and 12 weeks old chickens, respectively. He also reported only one value of heritability at 16 weeks of age (0.47). Gowe et al (1973) estimated heritabilities of 0.24 to 0.74 with a mean of 0.55 in layer type hens. Segura et al (1990) gave heritabilities of 0.51 and 0.86 for body weight of 8 months old roosters from a population selected for economically important egg traits and a control line, respectively. Hernandez and Segura (1994) using another population of Creole chickens in Yucatan, Mexico estimated heritabilities of 0.87 and 0.51 for birth weight, using sire and sire plus dam components of variance, respectively. Furthermore, Chambers (1990) notified that heritabilities for body weight of broilers tend to increase with age. However, in this study heritabilities for the early growth traits of the Creole chickens showed a non linear trend. Differences in heritability estimates could be attributed to method of estimation, breed, environmental effects and sampling error due to small data set or sample size. Environmental (high temperature and humidity) and poor management conditions, are known to increase the residual variance and decrease the heritability estimates. 

Table 1. Estimates of direct (h2) and maternal (m2) heritability and direct-maternal genetic correlation (ram) and fractions of variance due to common environmental (c2) effects and total phenotypic variance () for body weights of Creole chickens in Yucatan, Mexico.

Body weight:











4 weeks






8 weeks






12 weeks






16 weeks






Animal models by DFREML program (Meyer 1998).

The variance due to maternal additive and common environmental effects of dam disappears gradually as the chicks grow older. This means that a simple animal model is appropriated for genetic parameter estimates after 8 weeks of age in Creole chickens. The definition of the correct model is important, because the more complex the model, the larger the time needed for solution. This is even more important with large amount of data and in multi-trait analysis, because CPU time is a function of the number of variance and covariance components to be estimated.  

Maternal environment affects bird growth in two stages, namely the preovopositional maternal effect and the postovipositional maternal effect. The postovipositional effect can be divided into prehatch (incubation) and posthatch effects. The posthatch maternal influence on chick growth was not important in this study because birds were raised independently of the dams. Therefore, the common environmental effects that may possibly affect chick growth are preovipositional maternal components, such as egg size, egg weight, shell quality and yolk composition (Aggrey and Cheng 1993). The c2 estimates include influence of the oviductal environment, nonadditive gene action, and any sire-dam interaction that may be present. In broilers, Tullett and Burton (1982) observed that 97% of the variation in chick weight at hatch can be explained by two factors: fresh egg weight and weight loss during incubation. Pinchasov (1991) observed that the initial high correlation between egg weight and hatch weight declined with age. North (1986) indicates that the weight of the chick represents approximately 70% of the egg weight. It is therefore desirable to separate common environmental effects from heritability estimates, in order to better predict response to selection. Meyer (1992) addressed the difficulty of statistically separating the direct and maternal components of variance. Notter and Hough (1997), in sheep, suggested that partitioning maternal effects into additive and environmental components requires large amount of data and the presence of related dams.  

The antagonism between the additive direct-maternal effects found in this study for birth weight (-0.15) have been observed in other domestic species (Tosh and Kemp 1994, Diop and Van Vleck 1998). 

Finally, the parameters here estimated for Creole chickens under an enclosed artificial environment may be different under scavenging, free-run environment conditions, if genotype x environment interaction were important. In other words, parameter estimates for Creole chickens must be estimated if genetic improvement programs are planned for chickens under scavenging conditions. 

In conclusion, this study showed that additive heritabilities are low for body weight of Creole chickens during rearing. It also showed that direct maternal and common environmental are not important sources of variation after 4 weeks of age in Creole chickens. 


The authors wish to thank CONACYT for the financial support for this research (Project 31664-B).


Aggrey S E and Cheng K M 1993 Genetic and posthatch parental influences on growth of pigeons squabs. Journal of Heredity 84: 184-187 

Chambers J R 1990 Genetics of growth and meat production in chickens. Pages 559-643. In: Poultry Breeding and Genetics. R D Crawford editor. Elsevier, Amsterdam, The Netherlands.  

Diop M and Van Vleck L D 1998 Estimates of genetic parameters for growth traits of Gobra cattle. Journal of Animal Science 66: 349-355 

Falconer D S 1981 Introduction to Quantitative Genetics 2nd edition Longman, London and New York 

García E 1988 Modificaciones al Sistema de Clasificación Climática de Koeppen. Instituto de Geografía, Universidad Nacional Autónoma de México. México, D F  276p. 

Gowe R S, Lentz W E and Strain J H 1973 Long term selection for egg production in several strains of White Leghorns: Performance of selected and control strains including genetic parameters of two control strains. Proceedings 4th European Poultry Conference London, England, U.K. pp 225-245 

Hernández E y Segura J C 1994 Estimación del índice de herencia para peso al nacer en pollos criollos cuello desnudo Universidad y Ciencia 11(21-22): 121-124 

Kinney T R Jr 1969 A summary of reported estimates of heritabilities and of genetic and phenotypic correlations for traits of chickens. Handbook No. 363 Agricultural Research Service. 44 pp 

Meyer K 1992 Variance components due to direct and maternal effect for growth traits in Australian beef cattle. Livestock Production Science 31:179-188 

Meyer K 1998 DFREML Users Notes. University of New England, Armidale Australia 

North M 1986 Manual de Producción Avícola. 3ª. Edición editorial El Manual Moderno, México, D F 856 p 

Notter D R and Hough J D 1997 Genetic parameter estimates for growth and fleece characteristics in Targhee sheep. Journal of Animal Science 75: 1729-1737 

Pinchasov Y 1991 Relationship between weight of hatching eggs and subsequent early performance of broiler chicks British Poultry Science 32: 109-115 

Segura J C, Gavora J S, Fairfull R W, Gowe R S and Buckland R B 1990 Heritability estimates of male reproductive and morphological traits and their genetic correlations with female egg production and other related traits in chickens Poultry Science 69:493-501 

Segura-Correa J C and Juarez-Caratachea A 2002 Growth until 18 weeks of age of Creole chickens raised under intensive management conditions in Yucatan, Mexico British Poultry Science (in press) 

Tosh J J and Kemp R A 1994 Estimation of variance components for lamb weights in three sheep populations Journal of Animal Science 72: 1184-1190 

Tullett S G and Burton G F 1982 Factors affecting the weight and water status of the chick at hatch British Poultry Science 23: 361-369. 


Received 27 June 2002, accepted 8 November 2002

Go to top