Livestock Research for Rural Development 24 (6) 2012 Guide for preparation of papers LRRD Newsletter

Citation of this paper

Estimates of genetic and phenotypic parameters for milk traits in Arsi-Bale goat in Ethiopia

Mohammed Bedhane, Aynalem Haile, Hailu Dadi and Tesfaye Alemu T

Adami Tulu Agricultural Research Center, P.O.Box 35, Ziway, Ethiopia
benegash@gmail.com

Abstract

This study was conducted to evaluate milk yield performance and to estimate the genetic parameters for milk yield traits in Arsi-Bale goats.  A total of 227 records of milk production traits were used. Fixed effects were found to be important source of variation for most of studied traits.

 

The estimated performance for milk production traits were 86 days of lactation length (LL), 17.73kg of lactation milk yield (LMY) and 208.72g of daily milk yield (DMY). Based on the developed models, the obtained direct additive heritabilities (h2a) for LL, LMY and DMY traits varied from 0.00 to 0.03, 0.00 to 0.69 and 0.02 to 0.071, respectively. The heritabilities of maternal genetic effect (h2m) for LL, LMY and DMY were 0.03 ± 0.15, 0.22 ± 0.12, and 0.26 ± 0.12, respectively. The estimates of repeatabilities for LL, LMY and DMY were 0.08, 0.42 and 0.48 respectively. Except for LL, the studied traits have low to high estimates of heritabilities and correlation. Even though repeated measurements of the traits are needed, the current result can be used for Arsi-Bale goats genetic improvement programme. Fixed effects were found to be very important and they should be included in the intended genetic improvement programme.

Key words: Arsi-Bale goats, Genetic parameters, Milk yield


Introduction

Goats are one of the most important domestic animals in the tropics. They have a variety of functions, and in comparison with other ruminants, they possess unique abilities to adapt and maintain themselves in harsh tropical environments. Goats are multipurpose animals producing meat, milk, skin and hair. They occupy an important niche in the smallholder production system mainly due to low initial capital requirement, ability to produce food and fiber at relatively low cost, production of milk and meat in readily useable quantities, relatively high rate of reproduction potential, and marketability in a short period (Kevin and Wilson 2000). In tropical countries, as a result of increased human population and reduced land size, it is becoming difficult to meet all domestic milk demand by increasing cow milk production. This led to an increasing interest in goat milk production with many development programmes now in operation in the developing countries (Girma 1996).  According to the recent survey report of Tesfaye et al (2011), the farmers at low land part of the country (Ethiopia) largely practices multiple breeding objectives including milk production, growth, body size, and reproduction.

The estimated goat population size in Ethiopia is 22.8 million (CSA 2011). Out of this, approximately 27 percent are found in crop-livestock mixed farms in the highlands and the rest usually inhabit in arid and semi-arid lowland areas. The indigenous goats breeds are comparatively worthy for some traits like heat tolerance, disease resistance and ability to withstand poor management conditions. These indigenous breeds thrive and produce on sloppy, marginal and often uncultivable extremely high and low lands. Unfortunately, a large number of these genetic resources have been lost due to uncontrolled traditional breeding practices and lack of breed improvement programmes (Philipsson et al 2006). Any endeavor designed to correct the risk of uncontrolled traditional breeding practices and boost the productivity of goats should largely depend on effective genetic improvement programmes. The potential for genetic improvement in economically important traits of goats in a selection programme depends on the extent of the genetic variation and estimates of the genetic and phenotypic parameters. With this background, the present study was carried out to assess the genetic and phenotypic parameters of milk production traits in the Arsi-Bale goats.


Materials and Methods

Location and management

The Adami Tulu Agricultural Research Center (ATARC) is located in Mid Rift Valley of Ethiopia. The center is situated at an altitude of 1650 meter above sea level. It lies at latitude of 7°9'.N and 38°7'.E longitude. The study area has hot and semi-arid climate and it receives bimodal unevenly distributed average annual rainfall of 760.9 mm. The pattern of annual rainfall distribution in the area is categorized as dry period: December - February; short rain: March-May; main rain: June-August; and early dry: September-November (ATARC 1998). According to this report, vegetation coverage of the area is acacia woodland type and dominated by pennesitum and cinchrus grass species. 

The flock management system was semi-intensive type. Goats browsed/grazed outdoors throughout the year during the day time (08:00 am to 05:00 pm) with the access to water two times a day and were housed during the night time in pens with half wooden walls and corrugated iron roof and concrete floor. The doe and buck flocks were separately herded during the day time and mating was practiced only at night time throughout the year with 15-20 does to one buck ratio.   

There was no any especial management for milking does group during milking day except they were kept separately from their kids over the night in closed conventional house. The test day milking sampling was done once a week at morning only. Frequency of milking times in this collected data was only once per day and during milking, kids were separated from their mothers’ throughout the  day and night (24 hours) then, on the next day at the morning, direct or hand milking was practiced and measured immediately using scaled balance in gram. 

 

Traits and Fixed effects

Data used in this study were collected for Arsi-Bale goats from 2001 - 2006 for milk production traits. The data were collected from various records and entered into a computer database. Data validation and all necessary editions were carried out before making any analyses. A total of 227 records were used for milk yield traits. Additionally, lactation milk yield (LMY) generated from the existing data. A total of 21 sires and 55 dams were used for the estimates of genetic parameters for milk production traits. Daily milk yield (DMY) greater than zero and lactation length (LL) greater than sixty days was taken as a normal and used in the analysis. These truncation points were fixed on the basis of Muller (2005) in British Alpine, Saanen and Toggenburg dairy goats in South Africa. The fixed effects data were recorded on the same sheet with milk data and some of them were extracted and calculated from the existing data. The considered fixed effects were year of kidding, season of kidding, and parity of the dam, with levels of six, four and six, respectively. 

 

Statistical analysis
 
Least squares means analysis was carried out to examine the influence of fixed effects using the Generalized Linear Model (GLM) procedure of SAS (SAS 2002).  In the preliminary analysis, interactions of fixed effects were not significant (P > 0.05) in all cases and excluded from the final model.

 

Model 1: for milk production traits:

Yijk = µ + Bi + Xj + Pk   + eijk

Where, Yijk = is the observational value of the milk yield traits of the ith year, jth season and   kth parity

 µ =estimated overall mean

 Bi = the fixed effect of ith year (i =2001-2006)

 X = the fixed effect of jth season (j = dry, early dry, short rain, main rain)

P = the fixed effect of kth parity (k = 1, 2, 3, 4, 5, 6+)

 eijk = random error associated with Yijkth observation

Important fixed effects for all milk production traits were identified from preliminary analysis, using the Generalized Linear Model (GLM) procedure of SAS (SAS 2002). The (Co) variance and genetic parameters were estimated using the Derivative-Free Restricted Maximum Likelihood (DFREML) computer package (Meyer 1998). The (Co) variance components and genetic parameters (heritability, repeatability, direct genetic and phenotypic correlations) were estimated by fitting an animal model which fits direct additive effects, maternal additive effect plus permanent environmental maternal effect due to repeated records as random effects and the significant fixed effects listed earlier.

The representations of the animal models used to estimate genetic parameters for milk production traits are as follows:

 

Y = Xb + Z1a + e                                                     (1)

 Y = Xb + Z1a + Z2m + e     Qam = 0                       (2)            

Y = Xb + Z1a + Z3p + e                                           (3)

 Y = Xb + Z1a + Z2m + Z3p + e         Qam = 0          (4)

Where Y is the vector of records, b is a vector of an overall mean and fixed effects with incidence matrix X; a, m, p are vectors of random additive direct genetic, additive maternal genetic, and permanent environmental maternal effects with incidence matrices Z1, Z2 and Z3 respectively, and e is a vector of random errors. The repeatability of the milk traits was calculated from the third model which contains permanent environmental maternal genetic effects while the genetic and phenotypic correlations were calculated from the full model (model 4).


Results

Fixed effects and phenotypic parameters

Analysis of variances for lactation length (LL), lactation milk yield (LMY) and daily milk yield (DMY) are indicated in Tables 1. In the current study, LL was significantly (P < 0.001) affected by kidding year, kidding season and parity. LMY was not influenced (P > 0.05) by kidding year and parity. On the other hand, kidding season had a significant (P < 0.01) effect on LMY. DMY was significantly (P < 0.001) affected by kidding year and season. Parity did not affect (P > 0.05) DMY. Least squares means for milk performance traits of lactation length (LL), lactation milk yield (LMY) and daily milk yield (DMY) are indicated in Table 2. In the current study, the performance of Arsi-Bale goats for milk production traits were 86 days of LL, 17.73kg of LMY and 208.72g of DMY.

 

Table 1: Analysis of variance for lactation length, lactation milk yield and daily milk yield

Source

Lactation length

Lactation milk yield

Daily milk yield

 

DF

MS

DF

MS

DF

MS

Kidding year

5

344**

5

72 ns

5

18615**

kidding season

3

619***

3

185**

3

32862***

Parity

5

270***

5

18 ns

5

1332 ns

Model

13

448***

13

71*

13

6164***

Error mean square

213

84

213

35

213

4767

R2%

24.6

11.1

16.8

CV%

10.56

31.8

31.9

*, P < 0.05; **, P < 0.01;  ***, P < 0.001; ns, non-signifiant

 

Table 2: Least squares means and standard error for effects of lactation number, kidding year and kidding season on lactation length (LL), lactation milk yield (LMY) and daily milk yield (DMY)

Effect

Level

n

LL(days)

n

LMY(kg)

n

DMY(gram)

Overall

-

227

86 ±  1

227

18 ±  0.9

227

209 ± 10.7

 Parity

1

71

90 ±2a

71

18 ± 1.1

71

204 ± 12.8

2

60

85 ± 2ba

60

17 ± 1.1

60

207.6 ± 12.9

3

51

83 ± 2b

51

17.5 ± 1.2

51

212.9 ± 13.9

4

26

90 ± 2a

26

19 ± 1.3

26

216 ± 15.3

5

13

83 ± 3b

13

18.5 ± 1.9

13

221.8± 22.4

≥6

6

84 ± 4ba

6

15.7 ±2.6

6

190 ± 31

Kidding

year

2001

28

80 ± 3b

28

21 ± 1.8

28

252 ± 21a

2002

49

81 ± 2b

49

19.4 ± 1

49

241 ± 15.8a

2003

76

87 ± 1ba

76

18.1 ± 0.9

76

213  ± 11a

2004

4

89 ± 5ba

4

14.6 ± 3.1

4

164.8 ± 35.8b

2005

4

89 ± 5ba

4

17.5 ± 3.1

4

203.8 ± 37ba

2006

66

90 ± 1a

66

15.8 ± 0.8

66

175.7 ± 9.8b

kidding Season

1

31

82 ± 2b

31

17 ± 1.3b

31

207.2 ± 15.6b

2

77

84 ± 2b

77

20.5 ± 1.2a

77

242.4 ± 14.3a

3

63

85 ± 2b

63

18.1 ± 1.3ba

63

214.2 ± 15.4ba

4

56

92 ± 2a

56

15.4 ± 1.2b

56

171.2 ± 14.1b

Least squares means with different letters within a column for the same effects differ (P < 0.05)

Genetic parameters

 

Estimates of (Co) variance components and genetic parameters for milk production traits of Arsi-Bale goats are presented in Table 3. The estimates of direct additive heritabilities (h2a) for LL were low and varied slightly from 0.00 to 0.03. The estimates of direct additive heritabilities (h2a) from the four animal models varied from 0.02 to 0.71 and 0.001 to 0.696 for DMY and LMY, respectively. In the current study the estimates of maternal additive heritability from model 2 and 4 for LL, LMY and DMY were 0.0001 and 0.03, 0.01 and 0.26, 0.01 and 0.22, respectively. In Arsi-Bale goat breed, the estimated maternal permanent environmental effect ( C2 ) were 0.00 and 0.001 for LL, 0.12 and 0.25 for LMY and 0.12 and 0.36 for DMY from model 3 and 4 respectively.  

 

Table 3: Estimates of (co)variance components and genetic parameters for milk production traits using multitrait animal model

Traits

Models

Parameters

 

 

σ2a

σ2m

σ2pe

σ2e

σ2p

h2 a

h2 m

C2= σ2pe/ σ2p

Log Likelihood

LL

Model 1

2.93

-

-

94.6

97.5

0.03 ± 0.12

-

-

-1015

Model 2

0.01

3.7

-

101

105

0.00 ± 0.21

0.03 ± 0.15

-

-1041

Model 3

0.40

-

0.13

98.8

99

0.01 ± 0.18

-

0.00 ± 0.15

-1082

Model 4

0.09

0.1

0.04

102.3

102.5

0.00 ± 0.19

0.00 ± 0.25

0.00 ± 0.27

-1064

LMY

Model 1

23.12

-

-

10.1

33.2

0.69± 0.07

-

-

-1015

Model 2

0.49

7.36

-

20.9

28.7

0.02 ± 0.12

0.26 ± 0.12

-

-1041

Model 3

Model 4

15.79

0.03

-

0.18

4.6

6.9

19.2

20.4

39.6

27.5

0.40 ± 0.75

0.00 ± 0.28

-

0.01± 0.39

0.12 ± 0.70

0.25 ± 0.35

-1082

-1064

DMY

Model 1

2747

-

-

1139

3885

0.71 ± 0.06

-

-

-1015

Model 2

72.3

791

-

2714

3577

0.02 ± 0.13

0.22 ± 0.12

-

-1041

Model 3

2422.9

-

611

1946

4980

0.49 ± 0.8

-

0.12 ± 0.75

-1082

Model 4

111.92

52.91

1308

217

3643

0.03 ± 0.32

0.01 ± 0.64

0.36 ± 0.63

-1064

σ2a, direct additive variance; σ2m, maternal additive genetic variance; σ2pe, maternal permanent environmental variance; σ2e, error variance; σ2p, phenotypic variance; h2 a, direct heritability; h2 m, maternal heritability and C2 =  σ2pe/ σ2p

The estimates of repeatabilities, genetic and phenotypic correlations in Arsi-Bale goat breed for LL, LMY and DMY are presented in Table 4. Owing to the relatively small data size used in the current study, most of the genetic correlations had relatively larger standard error and the standard error of repeatability and phenotypic correlation was not available by the time of data analysis. Based on model 3, the estimates of repeatability for LL, LMY and DMY were 0.08, 0.42 and 0.48, respectively. The genetic correlation between LL and LMY was 0.43 while LL was correlated with DMY negatively (-0.01). The phenotypic correlation between LL and LMY was 0.15. The phenotypic correlation between LL and DMY was negative (-0.19).

 

Table 4: Estimates of repeatability on the diagonal, direct genetic correlations (below diagonal) and phenotypic correlations (above diagonal) between Lactation length (LL), Lactation milk yield (LMY) and Daily milk yield (DMY) in Arsi-Bale goat

Traits

LL

LMY

DMY

LL

0.08

0.15

-0.19

LMY

0.43

0.42

0.49

DMY

-0.01

0.31

0.48

 

Discussion

 

Fixed effects

 

The influence of parity on daily and lactation milk yield traits in Arsi-Bale goats was not significant whereas the influence of parity is highly significant for Toggenburg dairy type goats raised in Eastern highland of Kenya (Ahuya et al 2009). The difference between the two breeds (Arsi-Bale and Toggenburg) is wide and difficult to compare each other, because Arsi-Bale goats have shown a tendency to be a meat type goat rather than dairy type like Toggenburg goats. This inconsistency arise may be due to breed difference between those dairy type breeds and Ethiopian indigenous goat breeds which are not selected for milk production traits. According to the recent report of Hamed et al (2009) on Zaraibi goats in Egypt, parity, season and year of kidding affected significantly the total milk yield where as in the current stud those fixed effects except kidding season did not affect the LMY of Arsi-Bale goats.

 

 In the current study, the highest DMY was recorded in the short rainy season. Naturally, goats prefer moderate dry season than extreme wet season. Generally, wet season is not conducive environment for goat because during wet season due to high rain fall and muddy ground, goats are limited for selective browsing. In addition, the loose or open housing system at ATARC has its own negative influence on goat performance due to the fact that animals expend more heat for the purpose of body temperature maintenance during wet or heavy rainy season. As a result, short rainy season and early dry seasons are favorable for better milk production in the study area (Mohammed 2010).

 
Milk yield performance

In the current study, the performance of Arsi-Bale goats for milk production traits were 86 days of LL, 17.73kg of LMY and 208.72g of DMY. Tatek et al (2004) reported 296.1g milk yield per day, 83.4 days of lactation length and 21.72kg of milk per lactation for Arsi-Bale goats from on-farm experiment. These slight yield variations in the current and previous study may arise from the management practice (on-farm versus on-station), differences in milking frequencies per day (unfortunately one time per day (after 24 hours) at on-station and two times per day at on-farm) and other management practices such as especial management for lactating does brought the variation.  Daily milk yield off takes of 250 to 500ml were recorded in the high lands at their third or above parity (Farm Africa 1996). According to the report of Girma (1996), supplementation of high energy diet for lactating does produced on average 11.5% more milk than controlled group in Somali goats. This experiment showed that milk yield is the most vulnerable trait to poor management condition.

 The estimates for milk production traits in current study were lower than the temperate goat breeds (Girma 1996; Knight and Garcia 1997; Güney et al 2006). Variation in milk yield performance among different goat breeds can be explained by difference in breed performance and management condition. In addition, such differences may also be related to difference in body size of the animal. Lactation milk yield in the indigenous breeds of the tropics and sub-tropics are lower than those in temperate breeds in temperate climate, which could yield as high as 489.4kg per lactation (Güney et al 2006). It is obvious that lactation milk yield depends on lactation length and seasonal breeders have long lactation period while tropical breeds (non-seasonal breeders) have short lactation period. In the current study, DMY was obtained by hand milking procedures rather than the efficient method which is called weight-suckling-weight. The report of Girma (1996) revealed that the efficiency of the weight-suckling-weight milking method is better than hand milking and in most cases residual milk sucked by kids exceeded that harvested by hand. On the average, hand milking removed only 38% of the total milk produced. The remaining milk i.e. 62% was obtained by kid suckling. These and other management conditions could be the main reasons for the underestimation of milk production in the current study (Mohammed 2010).

 

Genetic parameters

In the current study low heritability for LL was reported. Similarly, low heritability (0.06) for lactation length was reported in Saanen dairy goats (Ribeiro et al 2000). Sample size, breed type and environmental variation may bring this slight difference. In contrast to these results direct additive heritability (h2a ) of 0.32 was reported for LL in Alpine goat breed (Mourad 2001). In addition, Ahuya et al (2009) stated that poor data collection and data management resulting in relatively high level of environmental variation in the traits. Reports of Hamed et al (2009) indicated that selection for low heritable traits will take a long time. Estimates of high (h2a )  for milk yield traits were documented in many literatures for different goat breeds (Kala and Prakash 1990; Mavrogenis and Papachristoforou 2000). In the current study the estimates of   h2a for all traits (LL, DMY and LMY) under models 1 and 3 were higher than model 2 and 4. The inclusion and exclusion of maternal additive and permanent maternal environmental genetic effects in the model caused a wide variation among the estimated h2a in the current study. So, using simple animal model exaggerates the estimates. The wide variation of direct additive heritability for DMY and LMY in Arsi-Bale goat breed indicates that there is highly environmental variation or variable management condition at the studying location.

 The current findings were in agreement with other report (Weppert and Hayes 2004). In many research works related to milk production traits, the estimates of  maternal heritability is very low and most of the time  maternal genetic effect was not included in the models but in the current study, the inclusion and exclusion of additive maternal genetic effects creates a wide variation in estimate of direct additive heritability for milk production traits. Accurate estimates of the genetic parameters largely depend on the pedigree structure, unconsidered fixed effects, sample size, creating large environmental variations and model selection (Zhang et al 2009). Based on the current estimates, permanent maternal environmental effect (C2) was less important for LL than DMY and LMY. These findings are in agreement with the C2 estimates of 0.05 for lactation length and 0.42 for milk yield in Spanish Assaf (Assaf.E) dairy sheep (Gutierrez et al 2007).  

 Generally, genetic differences between breeds in their phenotypic appearance and their production characteristics have arisen because different breeds have usually been developed in different localities and breeders with different aims. Differences in heritability estimates for any one trait can arise because genetic variability may differ from one breed or population to another or because the heritability estimates are derived from animals kept in different environmental conditions.

 
Repeatability and correlation

In the current study, repeaetabilities (0.08, 0.42 and 0.48) for LL, LMY and DMY, respectively were comparable to the repeatability reported for various goat breeds (Ilahi et al 2000; Ribeiro et al 2000). In addition, Hamed et al (2009) reported repeatability of 0.30 for total milk yield of Zaraibi goats in Egypt. Estimated repeatability’s value in this study was higher than the corresponding heritability. Repeatability is needed for any prediction of probable performance because it is often of interest to use information early in the productive life of a doe, to predict how productive she will be over the lifetime. According to a research work done in South Africa by Olivier (2005), in dairy goats, lactation yield is a primary trait for which repeatability is important. Repeatability is not a fixed value; it varies from herd to herd, and from one environment to another. Factors that affect heritability tend to affect repeatability in a similar manner. According to Hamed et al (2009), an estimate of heritability and repeatability of traits is important in breeding programme, because it determines at what stage the selection would affect the subsequent herd performance.

 The genetic correlation between LL and LMY was 0.43 while LL was correlated with DMY negatively (-0.01). The positive relationship show that both traits can be improved at the same time in genetic improvement program while the negative relationship indicated that care should be taken on the antagonist traits during selection programme. The phenotypic correlation between LL and LMY was 0.15. The association was positive and low. A research done in Kenya by Ahuya et al (2009) reported the phenotypic correlation between LL and LMY of Toggenburg dairy goat breed to be 0.11 and it is in agreement with the current result. The phenotypic correlation between LL and DMY was negative (-0.19). tn the current study, the negative correlations of some milk yield traits were not as such critical because their strength were very low. Strong genetic correlation (0.82) between lactation length and milk yield and phenotypic correlation (0.21) between lactation length and milk yield was reported by Gutierrez et al (2007). The report also revealed that there is high genetic correlation (0.99) between test day milk yield and total milk yield in the Spanish Assaf (Assaf.E) dairy sheep breed. Accurate estimates of genetic parameters are vital for genetic improvement in livestock. Although, large data set is required for more accurate parameter estimates, the results obtained in this investigation can be applied for Arsi-Bale goats genetic improvement programme (Mohammed 2010).


Conclusion


Acknowledgements

The authors wish to acknowledge the Ethiopia Institute of Agricultural Research (EIAR) for funding the study. We also acknowledge all Adami Tulu Agricultural Research Center (ATARC) staffs for their contribution.


Reference

Adami Tulu Agricultural Research Center (ATARC) 1998 Thirty years of research experience. Oromia Agricultural Research Coordination service. Bulletin no 1. 42p. 

Ahuya C O, Ojango J M K, Mosi  R O, Peacock  C P and Okeyo A M 2009 Performance of Toggenburg dairy goats in smallholder production systems of the eastern highlands of Kenya, Small Ruminant Research 83, 7-13.  

Central Statistics Authority (CSA) 2011 Agricultural Sample survey volume II: Report on Livestock and Livestock characterstics in Ethiopia. Statistical Bulletine 505, February 2011, Addis Ababa, Ethiopia. 

FARM Africa 1996  Goat types of Ethiopia and Ertirea. Physical description and  management systems. Published by FARM Africa, London, UK, and ILRI (international livestock research institute), Nairobi, Kenya, 76p. 

Girma A 1996 Studies on performance potential of Somali goats and their crosses with Anglo-Nubian: A contribution to breed documentation and evaluation. The Verlag Dr. Koster press, Berlin. 151p.  

Guney O, Torun, O, Ozuyanik O and Darcan N 2006 Milk production, reproductive and growth performances of Damascus goats under northern Cyprus conditions, Small Ruminant Research 65, 176-179. 

Gutiérrez   J P , Legaz  E  and Goyache  F 2007 Genetic parameters affecting 180-days standardised milk yield, test-day milk yield and lactation length in Spanish Assaf (Assaf.E) dairy sheep, Small Ruminant Research 70, 233–238. 

Hamed  A, Mabrouk  M M, Shaat  I and Bata  S 2009 Estimation of genetic parameters and some non-Genetic factors for litter size at birth and weaning and milk yield traits in zaraibi goats, Egyptian Journal of Sheep and Goat science 4(2), 55-64. 

Ilahi  H, Manfredi  E, Chastin  P, Monod  F, Elsen  JM and Le Roy  P 2000 Genetic variability in milking speed of dairy goats, Genetics Research 75, 315–319. 

Kala  S N and Prakash  B 1990 Genetic and phenotypic parameters of milk yield and milk composition in two Indian goat breeds, Small Ruminant Research, 3, 475–484. 

Keiven  G P and Wilson  R P 2000 Introduction to Animal Science. John Wiley and sons, Inc. New York. 562p. 

Knights  M and Garcia  G W 1997 The status and characteristics of the goats (Capra hircus) and its potential role as a significant milk producer in tropics: A Review, Small Ruminant Research 26, 203-215.  

Mavrogenis  A P and Papachristoforou  C 2000 Genetic and phenotypic relationships between milk production and body weight in Chios sheep and Damascus goats, Livestock Production Science 67, 81–87. 

Meyer  K 1998 DFREML: Programs to estimate variance components by restricted maximum likelihood using a derivative-free algorithm, User notes, Ver. 3.0 

Mohammed Negash 2010 Estimates of genetic and phenotypic parameters for growth and milk traits in Arsi-Bale goats.  (unpublished MSc thesis, Haramaya University). 

Mourad  M  2001 Estimation of repeatability of milk yield and reproductive traits of Alpine goats under an intensive system of production in Egypt, Small Ruminant Research  42, 1- 4. 

Muller  C J C 2005 Genetic parameter estimation and breeding plans for the South Africa dairy goat herd. A PhD. Dissertation, Department of Animal Science, Faculty of Agriculture and forestry sciences, University of Stellenbosch, South Africa. 133p. (http://www.etd.Sun.ac.za/jspui/bitstream/10019/1126/1/Muller) 

Olivier  J J, Cloete  S W P, Schoeman  S J and Muller  C J C 2005 Performance testing and recording in meat and dairy goats, Small Ruminant Research 60, 83–93. 

Philipsson J,  Rege  J E O and Okeyo A M  2006 Sustainable breeding programmes for tropical farming systems. Module 3, Animal Genetics Training Resource, CD Version 2, 2006, ILRI-SLU 

Ribeiro  A C, queiroz  S A, Lui  J F, Ribeiro  S D A and Resende  K T 2000 Genetic and phenotypic parameters estimates of production traits and genetic trend of milk yield of Saanen goats in Southeast of Brazil, ARS VETERINARIA, 16(3), 198-203. 

SAS 2002 User’s Guide, Statistical Analysis System Institute Inc., Cary, NC USA. 

Tatek  W, Hailu  D, Mieso G and Dadi G 2004 Productivity of Arsi Bale goat types under farmers’ management condition: a case of Arsi Negelle. Proceedings of the 13th annual conference of the Ethiopian Society of Animal Production (ESAP) held in Addis Ababa, Ethiopia, 25-27 August 2004. 

Tesfaye K  Aynalem H  and Hailu D 2011 Smallholder goat breeding and flock management practices in the central rift valley of Ethiopia. Tropical Animal Health and Production. Trop Anim Health Prod DOI 10.1007/s11250-011-0033-9. 

Weppert  M and Hayes  J F 2004 Direct genetic and maternal genetic influences on first lactation production in four breeds of dairy goats, Small Ruminant Research 52, 173–178. 

Zhang C-Y, Zhang Y, Xe  D-Q, Li  X  Su  J and  Yang  L-G 2009 Genetic and phenotypic parameter estimates for growth traits in Boer goat, Livestock Science 124, 66–71. 


Received 25 February 2012; Accepted 27 March 2012; Published 1 June 2012

Go to top