Livestock Research for Rural Development 20 (12) 2008 Guide for preparation of papers LRRD News

Citation of this paper

On-farm performance of Arsi-Bale goats in Ethiopia receiving different concentrate supplements

Mieso Guru, Girma Abebe*, A Goetsch**, Fayisa Hundessa, Abule Ebro and Berhanu Shelima 

Adami Tulu Agricultural Research Center, P. O. Box 35, Zeway, Ethiopia

*Ethiopia Sheep and Goat Productivity Improvement Program,P. O. Box 15566, Addis Ababa, Ethiopia

**American Institute for Goat Research, Langston University,P. O. Box 730, Langston, OK, 73050, USA


An experiment was conducted to compare effects of supplements based on different byproduct feedstuffs on on-farm performance of yearling Arsi-Bale goats in the Adami Tulu district of Ethiopia.  The study was conducted during the main growing season (wet season) at the time when most grazing land is cultivated.  A farmer research group (FRG) was formed in each of three villages.  Each FRG consisted of nine farm households, with eight contributing three animals and one providing six.  A barn with three pens was constructed at one farm in each village.  One or two animals from each farm were allocated to three supplementation treatments.  Animals received supplements and resided at night in the barn pens.  Supplements, offered at 2.5% of body weight, consisted of 50% wheat bran, 1% salt, and 49% noug cake (N), formaldehyde-treated noug cake (F-N), or linseed meal (L). 


Initial body weight was 14.5 (SE = 0.18 kg).  Average daily gain was greater for L than for N (P < 0.05) and F-N (P < 0.08) (100, 113, and 134 g/day for N, F-N, and L, respectively; SE = 6.6).  The difference between the increase in estimated animal value due to supplementation and supplement cost was 51.87, 61.1, and 79.75 Ethiopian birr per animal for N, F-N, and L, respectively. 


In conclusion, based on average daily gain and the greater concentration of metabolizable energy in linseed meal vs. noug cake, energy appeared relatively more limiting to performance than protein.  Supplementation of goats with available byproduct feedstuffs offers a means of achieving marketable body weight and profit with suboptimal grazing conditions.

Key words: daily gain, formaldehyde, linseed meal, noug cake, wet season


In rural areas of many countries such as Ethiopia, small ruminants are often not supplemented with concentrate feedstuffs, resulting in slow growth and relatively old age at market (Tsegahun et al 2000). This causes a large proportion of consumed nutrients to be used for maintenance rather than growth, with an associated low efficiency of production (Tolera et al 2000).  There are many reasons why such management practices to achieve fast growth at young ages are not employed, among which is the cost of supplemental feedstuffs (Tolera et al 2000).  However, in some cases farmers simply may not be aware of the potential benefits of preferred technologies, which could be adopted if first demonstrated on-farm (Tibbo 2000).


Byproducts are most common supplemental feedstuffs for small ruminants in rural areas of Ethiopia and in other developing countries of the world as well.  One such byproduct in Ethiopia is noug (Guizotia abyssinica) cake.  Noug cake has a moderate crude protein (CP) concentration, is relatively high in neutral detergent fiber (NDF), and CP is rapidly and thoroughly degraded in the rumen (Little et al 1986; Nuwanyakpa and Butterworth 1986; Nsahlai et al 1999).  In some settings treatment of oilseed cakes with formaldehyde to increase passage of intact protein to the small intestine is advantageous (Kumar and Walli 1994a,b).  Another byproduct in Ethiopia and more widely available around the world than noug cake is linseed (Linum asitatissimum) meal.  Linseed meal is also moderate in CP; however, protein is less thoroughly degraded in the rumen and the NDF concentration is lower compared with noug cake (ěrskov et al 1983; Cheeke 2005; Preston 2008).  Oilseed meals high or moderate in CP are typically mixed with other byproducts lower in CP such as wheat bran.


Ideally on-farm research to study and demonstrate attributes of supplementation includes a negative control treatment, entailing animals not receiving a supplement.  However, this is difficult to achieve under practical conditions, as farmers may not be willing to participate without the incentive of free supplement.  Thus, frequently an approach taken is to include one or more types of supplements with different ingredient composition, with at least one treatment being considered a ‘standard’ for comparison.  In addition to economic comparisons, based on characteristics of different supplement ingredients it may be possible to ascertain the nutritional condition most limiting to animal performance.  In this regard, the objective of this experiment was to compare effects of supplementation with wheat bran mixed with noug cake, formaldehyde-treated noug cake, or linseed meal on on-farm performance of yearling Arsi-Bale goats in the Adami Tulu district of Ethiopia.


Materials and methods 

Sites and animals


The experiment was conducted in July to September of 2007 during the main growing season (wet season) at the time when most land is under crops in three villages of the Adami Tulu district.  A Farmer Research Group (FRG) was formed in each village.  Initially there were five villages and FRG, but two were excluded because of animal sale before the experiment was completed.  FRG consisted of nine farm households, eight contributing three animals to the experiment and one providing six.  Farm households were selected based on animals available and interest and willingness to participate as directed.  Farmers were trained in the necessary activities before the experiment, along with frequent visits and continual monitoring by researchers during the experiment.


One lead farm household in each FRG was selected for construction of a simple barn.  Because of additional involvement of these lead farmers, they were allowed to contribute six animals rather than three.  Barns consisted of three 3 x 3 m pens with a supplement trough.  One or two Arsi-Bale yearling (based on dentition) intact male goats of each farmer were allocated to the three supplement treatments.  In the evening goats were divided into the three treatment groups and placed separately in the three pens for supplementation.  Two animals were removed from the experiment because of reasons unrelated to treatments or the experiment, resulting in a total number of 88.  Goats remained in pens until morning.  For approximately 8 h during the day goats were allowed to graze nearby pastures with alternating management by the different households of the FRG.




Supplements consisted of 50% wheat bran, 1% salt, and 49% noug cake (N), formaldehyde-treated noug cake (F-N), or linseed meal (L).  Formaldehyde treatment entailed application of formalin (40% formaldehyde) at a rate of 1 mL/100 g of noug cake.  The SSA Feeds:Sub-Saharan Africa Feed Information System (retrieved July 20, 2008, for feedstuff samples in Ethiopia lists concentrations (dry matter basis) of CP of 31.7, 29.2, and 16.1% and of metabolizable energy (ME) of 8.95, 10.71, and 9.37 MJ/kg for noug cake, linseed meal, and wheat middlings, respectively.  Thus, predicted concentrations (DM basis) in N and F-N were 23.6% CP and 9.08 MJ/kg ME and in L were 22.4% CP and 9.92 MJ/kg ME.  The feeding rate was 2.5% of the most recent body weight (BW) determination.  Supplements were completely consumed throughout the experiment.  The experiment lasted 60 days, preceded by a 15-day adaptation period.  Animals were weighed periodically, with periods 15, 10, 10, 10, 7, and 8 days in length.


Vegetation characterization


In order to characterize grazing conditions, a vegetation survey was undertaken in the middle of the experiment.  Three sampling sites were randomly selected in areas grazed by animals of each FRG.  Belt transects, 50 x 4 m, were laid out to assess both herbaceous and woody vegetation layers.  Species composition was determined based on frequency of occurrence using a wheel point apparatus, with the nearest plant within 30 cm recorded (Tidmarsh and Havenga 1955).  At each sampling site 300 point observations were recorded at 3-m intervals.  If no vegetation was present, ‘bare ground’ was recorded in order to evaluate plant density (Mentis 1984) and real rangeland conditions (Danckwerts and Teague 1989).  At each sample site the percentage basal cover, which refers to the area occupied at the intersection of the plant and the soil surface and relates to protection of soil from erosion, was calculated by dividing the number of strikes by the total number of points.  Only live woody plants, regardless of being single- or multi-stemmed, were recorded and numbers of individuals of each tree and shrub species were counted to estimate woody vegetation density per hectare.


Statistical analysis economic evaluation


Data were statistically analyzed with GLM procedures of SAS (1990).  The model consisted of treatment, FRG, treatment x FRG, and farmer (FRG).  Treatment x FRG was used to test significance of treatment and FRG.  Initial BW was used as a covariate for analysis of BW and average daily gain (ADG).  Means were separated by least significant difference when the overall treatment F value was significant (P < 0.05).


An economic evaluation was based on feedstuff prices of 1.2, 1.7, 2.0, and 1.5 Ethiopian birr (ETB)/kg for wheat bran, noug cake, linseed meal, and salt, respectively.  The cost of formaldehyde treatment was estimated at 0.3 ETB/kg noug cake.  At the end of the experiment animals were sold to the Modjo Modern Export Abattoir (MMEA; Modjo, Ethiopia).  The average dressing percentage for Arsi-Bale goats from previous studies at the Adami Tulu Agricultural Research Center is 52% and the price of goat meat at the time of animal sale was 30 ETB/kg.  Gross return from supplementation was based on this price and the estimated increase in carcass weight.  However, it should be noted that the actual price received for all animals from the MMEA was 8.75 ETB/kg live weight.  This premium price was received 1 wk after the experiment ended, subsequent to a field day attended by personnel of the MMEA.  The economic evaluation was based on change in estimated carcass weight during the experiment rather than live weight because of the uncertainty concerning initial animal value.


Results and discussion 



Most abundant grass species (Aristida adscensionis, Chloris pycnothrix, and Eragrostis tenuifolia) were annuals, which is not preferable for protection of soil from erosion and provides a general indication of declining range conditions (Table 1).  Aristida adscensionis is usually grazed when young but is unpalatable in late flowering and seeding stages.  This forage is of some importance for grazing in dry areas.  Chloris pycnothrix is a very common weed especially in dry gravelly soils but also in moister regions.  It is readily grazed by cattle though is of low productivity.  Eragrostis tenufolia is a palatable grass of low productivity that often invades over-grazed pastures.  Of the grass species present, only Digitaria milanjiana and Cynodon dactylon are perennials.  Digitaria milanjiana is an excellent grass for grazing found on black heavy soils and in arid areas (CADU 1974; Van Oudtshoom 1999).  Fairly high levels of bare ground at two of the villages and presence of considerable Solanum incanum plants at the other village also reflect deteriorating grazing conditions. The average basal cover of 5.0% in the study sites is less than 50% of an excellent basal cover of 12% (Amsalu and Baars 2002).  Two of the other woody plants noted, Acacia tortilis and Balanites aegyptica, are highly palatable browse species for goats (Teshome 2006).

Table 1.  Vegetation available for grazing by male Arsi-Bale goats in the Adami Tulu district of Ethiopia receiving different concentrate supplements


Village or farmer researchg




Herbaceous vegetation, mean %

   Chloris pycnothrix




   Cynodon dactylon




   Digitaria milianjiana




   Eragrostis tenuifolia




   Aristida adscensionis




   Crotolaria spinosa




   Bare ground




   Basal cover




Woody plants, mean plants/ha




   Acacia tortilis




   Balanites aegyptica




   Solanum incanum








Body weight and average daily gain


BW was similar among treatments at all times except day 60, when it was greater for L than for N (P < 0.05; Table 2).  In accordance, average daily gain was similar among treatments in all periods up to day 52.  >From day 53 to 60, ADG was considerably greater for L vs. N and F-N (P < 0.05).  It is unclear why ADG only in the last 7 days of the experiment differed among treatments,  Nonetheless, this difference resulted in greater ADG in the entire experiment for L than for N (P < 0.05) and F-N (P < 0.08).


Average daily gain in this experiment was relatively high compared with previous findings of supplementation studies with similar animals (Mieso et al 2005; Mieso and Prasad 2006).  Reasons for higher ADG of goats supplemented with L than N or F-N cannot be conclusively discerned from measures of this experiment.  However, the higher ME concentration in linseed meal than noug cake (20%) suggests that energy was most limiting to performance rather than protein.  The lack of significant difference in ADG between N and F-N implies that amino acid absorption was ample given the amount of absorbed energy.  Likewise, the higher amount of ruminally fermented energy for L than for N or F-N would have elevated ruminal microbial protein synthesis.

Table 2.  Effects of different concentrate supplements on body weight and average daily gain of yearling male Arsi-Bale goats in the Adami Tulu district of Ethiopia








Body weight, kg





   Day 0





   Day 15





   Day 25





   Day 35





   Day 45





   Day 52





   Day 60





Average daily gain, g





   Day 1-15





   Day 16-25





   Day 26-35





   Day 36-45





   Day 46-52





   Day 53-60





   Day 1-602





1N, F-N, and L = supplement offered at 2.5% body weight, consisting of 50% wheat bran, 1% salt, and 49% noug cake, formaldehyde-treated noug cake, and linseed meal, respectively.

2Average daily gain of F-N and L differed at P < 0.08.

a,bMeans in a row without a common superscript letter differ (P < 0.05).

Economic evaluation


As noted earlier, without a negative control treatment it is not possible to definitively evaluate economic returns from supplementation.  Although, from anecdotal evidence and BW and the appearance of other animals of the villages of the same age, BW suitable for market could not have been achieved without supplementation.  Based on change in BW and supplement cost, all supplements were profitable (Table 3).  The numerically greater ADG for F-N vs. N resulted in slightly greater profit for F-N, and profit was highest among treatments for L.

Table 3.  Effects of different concentrate supplements on costs and returns with yearling male Arsi-Bale goats in the Adami Tulu district of Ethiopia






Total supplement intake, kg/animal




Total body weight gain, kg/animal




Total carcass gain, kg/animal




Gross return, ETB/animal




Cost, ETB/animal




   Wheat bran




   Noug cake or linseed meal




   Formaldehyde treatment












Profit, ETB3




1N, F-N, and L = supplement offered at 2.5% body weight, consisting of 50% wheat bran, 1% salt, and 49% noug cake, formaldehyde-treated noug cake, and linseed meal, respectively.

2ETB = Ethiopian birr.

3Difference between gross return and total cost.




This experiment was supported by the U.S. Agency for International Development (USAID)/Ethiopia under Agreement Number 663-A-00-05-00441-00 for the project entitled Ethiopia Sheep and Goat Productivity Improvement Program.  Appreciation is expressed to Mr. Assefa H Silasse, Shimals Gizechow, Abe Barko and Dejene Teddesse for assistance in data collection.



Amsalu S and Baars R M T 2002 Grass composition and rangeland condition of the major grazing areas in the mid rift valley, Ethiopia. African Journal of Range and Forage Science 19: 161-166


CADU 1974 An illustrated guide to grasses of Ethiopia. Chilalo Agricultural Development Unit. Asella, Ethiopia.


Cheeke P R 2005 Protein sources. In: Applied Animal Nutrition:  Feeds and Feeding, 3rd Edition. Pearson Education, Inc., Upper Saddle River, NJ, pp. 91-144


Danckwerts J E and Teague W R 1989 Veld management in the Eastern Cape. Pretoria pasture research section. Department of Agriculture and Water Supply. Eastern Cape Region


Kumar V and Walli T K 1994a Effect of feeding urea treated wheat straw supplemented with HCHO treated GN cakes on growth performance of crossbred calves. Indian Journal of Animal Nutrition 11: 29-33


Kumar V and Walli T K 1994b Nitrogen solubility of untreated and treated cakes in different solvents. Indian Journal of Animal Science 64: 989-994


Little D A, Anderson F M and Durkin J W 1986 A note on the effect of supplementation with noug cake (Guizotia abyssinica) on the milk production of crossbred cows. In: Utilization of Agricultural By-products as Livestock Feeds in Africa, Little D A and Said A N (Editors), Workshop Proceedings. African Research Network for Agricultural Byproducts (ARNAB). pp. 136-141. Retrieved July 20, 2008,


Mentis M T 1984 Monitoring in South African grasslands. Council for Scientific and Industrial Research. National Scientific Program Unit, Pretoria. South African National Scientific Program Report Number 91


Mieso G, Hailu D, Tasfaye L, Amsalu S and Esayas A 2005 Growth performance of Arsi-Bale goat kids supplemented with different levels of concentrate. Journal of Livestock Extension, Ibadan, Nigeria 3: 82-86


Mieso G and Prasad S 2006 Effect of energy and bypass protein supplementation on growth performance of crossbred kids. Indian Journal of Dairy Science 59: 229-232


Nsahlai I V, Umunna N N and Osuji P O 1999 Influence of feeding sheep on oilseed cake following the consumption of tanniferous feeds. Livestock Production Science 60: 59-69


Nuwanyakpa M and Butterworth M 1986 Effects of urea, molasses, molasses-urea, noug cake and legume hay on the intake and digestibility of teff straw by highland sheep. In: Utilization of Agricultural By-Products as Livestock Feeds in Africa, Little D A and Said A N (Editors). Workshop Proceedings. African Research Network for Agricultural Byproducts (ARNAB). pp. 87-98. Retrieved July 20, 2008,


ěrskov E R, Hughes-Jones M and Elimam M E 1983 Studies on degradation and outflow rate of protein supplements in the rumen of sheep and cattle. Livestock Production Science 10: 17-24


Preston R L 2008 Typical composition of feeds for cattle and sheep. In: Beef Magazine, Minneapolis, MN. Retrieved July 20, 2008,


SAS 1990 SAS User’s Guide, Version 6, 4th Edition, Volume 2.  SAS Inst. Inc., Cary, NC


Teshome A 2006 Traditional Utilization Practices and Condition Assessment of the Rangelands in Rayitu District of Bale Zone, Ethiopia. MS Thesis, Haramaya University, Alemaya, Ethiopia


Tibbo M 2000 Livestock production constraints in aa M2-2 sub-agroecological zone with special reference to goat production. In: The Opportunities and Challenges of Enhancing Goat Production in East Africa, Merkel R C, Abebe G and Goetsch A L (Editors), Conference Proceedings. pp. 92-106. E (Kika) de la Garza Institute for Goat Research, Langston University, Langston, Oklahoma, USA. Available at


Tidmarsh C E M and Havenga C M 1955 The wheel points methods of survey and measurement of semi-open grasslands and Karoo vegetation in South Africa. Memories of the Botanical Survey. Journal of Grassland Society of South Africa 7: 52-61


Tolera A, Merkel R C, Goetsch A L, Sahlu T and Negesse T 2000 Nutritional constraints and future prospects for goat production in East Africa. In: The Opportunities and Challenges of Enhancing Goat Production in East Africa, Merkel R C, Abebe G and Goetsch A L (Editors), Conference Proceedings. pp. 43-57. E (Kika) de la Garza Institute for Goat Research, Langston University, Langston, Oklahoma, USA. Available at


Tsegahun A, Lemma S, Sebsibe A, Mekoya A and Sileshi Z 2000 National goat research strategy in Ethiopia. In: The Opportunities and Challenges of Enhancing Goat Production in East Africa, Merkel R C, Abebe G and Goetsch A L (Editors), Conference Proceedings. pp. 1-5. E (Kika) de la Garza Institute for Goat Research, Langston University, Langston, Oklahoma, USA. Available at


Van Oudtshoom F 1999 Guide to grasses of Southern Africa. Briza Publications, Pretoria, South Africa

Received 10 September 2008; Accepted 9 October 2008; Published 5 December 2008

Go to top