Livestock Research for Rural Development 15 (2) 2003

Citation of this paper

Repeatability of litter traits of Nigerian indigenous sows


A A Adeoye, I O Adeogun and J O Akinokun


Department. of Animal Science, Obafemi Awolowo University, Ile –Ife, Nigeria


Records of 252 litters of Nigerian indigenous pigs sired by 48 boars and 83 sows over a period of 11 years (1977-1987) at the swine Unit of Obafemi Awolowo University Teaching and Research Farm were used to determine the repeatability estimates of litter trails. Data were collected on litter size, litter weight and average pig weight at birth and at weaning. Record for each trait were pooled and analysed to estimate the repeatability. Repeatability estimates for litter size at birth, litter weight at weaning, pig weight at birth, pig weight at weaning, litter size at weaning and litter weight at birth respectively, were as follows: 0.27+ 0.07; 0.12 + 0.07, - 0.025 + 0.06, 0.19 + 0.07, 0.18+0.006, and 0.11+0.007. It is concluded that for genetic improvement of litter traits of Nigerian indigenous sow under the existing management system, two or more records will be required to estimate the breeding value of individuals in the herd.


Key words:  Litter traits, Nigeria, repeatability, sows,



Low production potential of Nigeria indigenous pigs have been reported by many researchers (Adebambo 1986; Chiboka 1981), though superior adaptation of this animal to different management systems in the tropics compared to the improved exotic breeds were reported by Komolafe et al (1980). Pig breeds in developed countries produce lean meat under improved feeding and management condition. However, introduction of such genotype to all systems of production in developing countries is not practical owing to management and disease problems. 


Therefore, gene flow method that have been utilized in developed countries to replace indigenous population with superior genotype (Cunningham 1985) cannot be applied to improve all the indigenous pig breeds in developing countries because changes in the population structure might constitute a threat to the pig population causing a significant reduction in valuable genes (Polge 1983). 


Hence genetic improvement of indigenous breed have to be carried out under existing management and feeding system in order to produce suitable genotypes for the different production systems. However, a prerequisite for genetic improvement of stocks is the knowledge of genetic parameters (Repeatability, heritability and phenotypic and genetic correlations) of economic traits. 


The objective of this study was to estimate the repeatabilities  of litter traits of Nigerian indigenous sows. 

Materials and Methods


The animals for this study originated from Nigerian indigenous pigs maintained at the Swine Unit of Obafemi Awolowo University Teaching and Research Farm: 1,512 piglets from 252 farrowing, obtained from breeding 48 boars to 83 sows, for a period of 11 years (1977-1987) were used for this study. The feeding, breeding and management of the breeding stock were as reported by Llori et al (1984) and Chiboka (1981). 

Data were collected on litter size, litter weight and average pig weight at birth and at weaning. 


Statistical analysis 


The statistical model used for repeatability estimate was:

Ykm  = + a k  + ekm

Where:  Ykm  = measurement on the nth litters by kth sows
   = common mean.
 a = The effect of the Kth sows
             ekm  = The environment  deviation of the mth measurement within an individual 

Data for each trait were subjected to analysis of variance to estimate repeatability by method of Becker (1984) using SAS (1988). The standard error was calculated using the formula of Swigal et al (1964) for unequal number of measurements per sow. The graph of gain in accuracy for each trait was plotted by using formulae described by Falconer (1986). 

Result and Discussion


The repeatability estimates for the traits considered are shown in Table I.  

Table 1: Repeatability estimates of litter traits of Nigerian Indigenous sows



Litter size at birth

0.27.+ 0.07

Litter size at weaning

0.18 + 0.006

Litter weight at birth

0.11 + 0.007

Litter weight at weaning

0.12 + 0.7

Pig weight at birth

-0.025 + 0.006

Pig weight at weaning

0.19+ 0.07


The graphs of gain in accuracy for the traits are shown in Figure I.

Figure 1. Gain in accuracy for litter traits of pigs


The estimates for the traits were low, and there was negative estimate for pig weight at birth.


The value obtained for litter size at birth was higher than the values reported by other researchers (Jayarajan and Ulaganathan 1993; Ehiobu and Kyado 2000; Babot et al 1994). This may be due to reasons such as differences in breed and the environment condition under which the animals were kept (Falconer 1986).


The estimate for litter size at weaning was higher than the 0.13+0.55 value reported by Ehiobu and Keyado (2000). Similarly the repeatability for pig weight at weaning was lower than that estimated for pig weight at weaning in that study. The value obtained for pig weight at birth is lower than 0.12+0.08 reported by Ehiobu and Kyado (2000). The negative value may be due to sampling error.


The values reported for litter weight at birth and litter weight at weaning were lower than those reported by Dalton (1985).


The graphs of gain in accuracy for all the traits fell off rapidly as the number of measurements increased in all the traits considered. This is a characteristic of low repeatability estimate. The lower the estimate the more rapidly is the fall. Thus the fall was more in pig weight at birth and least in litter size at birth.


It could be concluded from the present study that the repeatability estimates of litter traits of Nigerian indigenous sow were comparable to those of improved exotic pigs in respect of the low values. Thus more than two records would be required to evaluate the breeding value of the individual to improve the herd performance.





Adebambo O A 1986 Selective breeding of the Nigerian indigenous Pigs for the rural producer. Proceedings 3rd World Congress on genetic application to Livestock Production Nebraska, USA: X: 70-74.


Babot D, Noguera J L, Actonso L and Estang G 1994 Influence of management effect and group size on the reproduction of breeding on genetic applied to livestock production. University of Guelph, Canada. 7-12 August 1994. 17: 35 – 54.


Becker W A 1984 Manual of procedures in quantitative genetics. 4th edition Washington State University Pullman Washington P. 62.


Chiboka O O 1981 The effect of age at first mating on litter characteristics in the native Nigerian pig. Livestock Production Science. 8(2): 155-159.


Cunningham G P 1985 Modern development in animal breeding. In Proeedings. 15th World Conference on Animal Production Tokyo. 14-19 August 1983 1: 35-42.


Dalton D C 1985; An introduction to practical animal breeding. ELBS (editor). P. 166-200.


Ehiobu and Kyado J A 2000 Performance characteristics of sow of Large White Hampshire and their crosses. Book of proceedings, 25th annual NSAP conference. 19-23 March Umudyal Pp. 260-261.


Falconer D S 1986 Introduction to quantitative genetics ELBS (Editor). Longman London. P 126-132


Llori J O, Yusuph W H and Somacle R 1984 The effect of dietary protein and palm oil on performance, age, and weight at puberty of Indigenous pigs in Nigeria. Ife Journal of Agriculture 6:54-64.


Jayarajan S and Ulaganathan V 1993 Studies on repeatability of litter traits of LW. Indian Veterinary Journal 70(8):  730-732.


Komolafe M F, Adegbola A A, Are L H and Ashaye T I 1980 Agricultural Science for West African schools and colleges. 2nd edition Oxford University Press Ibadan. Pp 191-196.


Polge C 1983 Plenary session 1. summary and conclusion in proceedings 5th World Conference Animal Production Tokyo, 14-19 august 1982 Volume 1 pp 79-82 cited in World Animal. Review 1987 p 3.


SAS 1988 SAS users guide. STATISTICS 1988 Ed. SAS Inc; Cary North Carolina 11:575-607 .


Swigal L A, Harvey W R, Everson O O and Gregory K E 1964 The variance of intraclass correlation involving groups with one observation Biometrics 28:818-826.



Received 31 January 2003; Accepted 7 February 2003

Go to top