Livestock Research for Rural Development 28 (4) 2016 Guide for preparation of papers LRRD Newsletter

Citation of this paper

Performance of scavenging ecotypes of indigenous chickens on targeted phase supplementary feeding

J K Gakige, A M King’ori, B O Bebe and A K Kahi

Department of Animal Sciences, Egerton University, P.O. Box 536-20115, Egerton, Kenya


Indigenous chickens (IC) ecotypes exhibit points of low and high growth rates along the growth curve, with a slowed initial phase followed by rapid exponential growth phase and a final slow growth phase. Weight gain varies depending on the growth stage in the scavenging ecotypes of IC, suggesting a possible variation in nutrients utilization between the different phases. Feeding interventions to improve productivity have been applied uniformly without considering the different growth phases. This study was designed to provide a scientific basis for feeding strategies that exploit the physiological adaptations induced by the requirements for improved growth rates. The objective was to determine the effect of supplementary feeding at different growth phases on mature weight, feed intake and feed conversion efficiency (FCE). Feeding trials were conducted on two ecotypes, Kakamega (KK) and Bondo (BN), for a period of 21 weeks with supplementation at three growth phases (5-8weeks, 9-14weeks and 15-21weeks) while leaving the chickens to scavenge for the rest of the growing period. The control group was left to scavenge for the entire study period.

The results showed a significant difference (p< 0.05) between the treatments where the highest mature body weights were recorded in 15-21weeks: 1021g for Kakamega (KK) and 1123g for Bondo (BN) ecotypes. However, FCE was highest at 9-14  weeks reaching 0.241 but the treatment group was not able to maintain high body weights after they were put on scavenging treatment. The study recommended that IC should be supplemented during the final growth phase (15-21 weeks) so as to achieve high body weights while minimizing the cost of supplementation. This feeding strategy can enable the IC resource poor farmers to achieve increased profits from keeping indigenous chicken.

Key words: feeding strategy, growth phase, nutrients, weight


Kenya has an estimated poultry population of over 31 million birds (FAO 2012). Out of these, 25 million are local or indigenous chickens, which are kept by 90% of rural communities (King’ori et al 2010) under free range conditions. Chickens from the same ecological zone are called ecotypes and are given names according to the regions (Ng’eno et al 2014). The origin of these ecotypes is as a result of adaptation, evolution, selection, mutation and genetic drift (Gondwe 2005). Msoffe et al 2001 reported distinct ecotypes in Tanzania. These populations have exhibited high between and within ecotype variation in body weight, egg weight, reproduction performance, plumage colour, comb type and skin colour. The same variation has been reported in Ethiopia (Tadelle and Ogle 2001), Zimbabwe (Mcainsh et al 2004), Botswana (Badubi et al 2006) and Kenya (Ng’eno 2011).

Due to the potential of the IC to improve rural livelihoods in Kenya (Magothe et al 2012), efforts to improve productivity through improved nutrition have been explored. However, many of these attempts have not succeeded but instead presented new challenges such as increased cost of production. Considering the high cost of supplementation in feeding indigenous chicken, there is need for feeding strategies that match the growth phases of the ecotypes. Ng’eno (2011) showed that growth curves of two IC ecotypes exhibited slowed initial phase followed by rapid exponential growth phase and a final slow growth phase, which suggests that supplementation at the different growth phases may have different effects on performance of the ecotypes.

Indigenous chickens (IC) are highly adaptable birds mainly reared under free-range conditions. The advantages of free range scavenging include availability of free feed resources in the surrounding environment and kitchen leftover, preserved ability to incubate and brood naturally (Swai et al 2005). However, their nutrient intake from the scavenging material is only sufficient for maintenance and low production (King’ori et al 2007). From scavenging, IC protein intake is 8.5g/day but the requirement is 11.7g/day (King'ori et al 2007). This leaves a deficit of 3.2g/day which requires supplementation. Supplementation with cereals is the most common practice, but this leads to competition for food with humans (Magothe et al 2012).

Materials and methods

The research was carried out at a poultry research unit of the Indigenous Chicken Improvement Program (InCIP), at Egerton University, Njoro Sub County, Nakuru County, Kenya. Njoro sub County lies on the longitude 35.9449 0 East of Greenwich Meridian and Latitude -0.33160 South of the equator. It is about 25km west of Nakuru town. The altitude is between 1800 and 2423 meters above sea level with a mean annual temperature of 17 - 22 0C and mean annual rainfall of 700 to 1200 mm range.

Management of experimental birds

Flocks were from a collection of eggs from two ecotypes of IC from Kakamega (KK) and Bondo (BN) which were chosen because there has been minimum exotic genetic dilution in these regions. One hundred and eighty eggs were collected from each area and each of the areas represents an ecotype sample

The eggs were numbered for tracking identification then artificially incubated. At hatching, each chick was weighed and wing tagged with an identification number. Brooding was from hatching to 5 weeks. All chicks were vaccinated against Marek’s disease at 1 day old, Gumboro disease at 18, 24 and 30 days old, Fowl Typhoid at 5 and 18 days old, Newcastle disease at 24 days, 6 weeks , 13 weeks and 15 weeks s old and Infectious Bronchitis at 6, 13 and 15 weeks old as per veterinarian’s recommendations. Any other incidence of disease condition was treated promptly by resident veterinarian. Age-weight data were recorded weekly for each bird until 21 weeks of age. Disinfection of brooding and rearing pens was done one week before introduction of the birds. Wood shavings were used as litter material at about 5 cm initial thickness.

Brooding of chicks was done in deep litter brooders fitted with infra-red electric bulbs. At the beginning of the 6th week, chicks were transferred to deep litter rearing pens (1m × 3m), within the same house. Each pen was provided with two feeders with a capacity of up to 10 kg of feed and two-5 liters water drinkers. The pens were designed to allow the chickens to move out and scavenge freely during the day time and get back to the pens in the evening. The chickens were allowed to scavenge together for uniformity of scavenging feeding resources then move back to their respective pens in the evening. There was group feeding where feed was given in each pen. Feed was offered at 8.00 hours where the chicken were allowed free access up to 10.00 hours and then released to scavenge. Feeding was done in the morning only as per the practice with most farmers.

Experimental design

Five weeks old chicks were subjected to a 2 x 3 x 4 factorial treatment arrangement with each of the two ecotypes under three growth phases assigned to four treatments. This was meant to capture the effect of treatment on the two ecotypes and the three growth phases.

The birds were offered a chick starter diet ad libitum from 0 to 5 weeks of age. Clean water was provided ad libitum daily. After 5 weeks of brooding; chicks were randomly allocated into four groups of twenty chicks for each ecotype. The first group, which was the control, was left to scavenge with no supplementation up to 21 weeks. The second group was supplemented from week 5 to 8 with a diet formulated to provide 160g CP/ kg and 12.56 MJ/kg ME then left to scavenge without supplementation up to week 21. The third group was left to scavenge without supplementation up to week 8 then supplemented with a diet containing 160g CP/kg and 10.89 MJ/kg ME up to 14 weeks then left to scavenge without supplementation up to week 21. The fourth group was left to scavenge without supplementation up to week 14 then supplemented up to week 21 with a diet containing 160g CP/kg and 10.05 MJ/kg ME. The variation in energy was as per the energy requirements at every growth stage (King’ori et al 2007).

Data collection

Feed intake, weight gain, FCEand mature weight were measured as follows; feed intake, which was the difference between feed given and feed refusal, was recorded daily, live weight per bird was recorded weekly, FCR was determined by calculating the ratio of feed taken to weight gained on a weekly basis and mature weight was measured at week 21. Regression analysis of the data was done to determine the effect of ecotype, feed intake and FCR on average daily weight gain of the KK and BN ecotypes.

Statistical analysis

The PROC GLM of SAS (1998) was used for analysis of variance of body weights at each age. The fitted model accounted for the fixed-effects of ecotype, treatment, phase and interaction between ecotype and treatment, ecotype and phase, phase and treatment, phase ecotype and treatment. Least square means were separated using the probability differences option. The model fitted was:

Yijk = µ + α i + β j +rk + ( αβ )ij+( α r)ik +( β r)jk +( αβ r)ijk + ε ijk

where: Yijkl = body weight of the ith bird at a particular age; μ=overall mean; αi =fixed effect of i th treatment (i=1, 2, 3, 4 treatments); βj =fixed effect of jth phase (j= 1, 2, 3 phases); rk = fixed effect of kth ecotype (k = KK and BN); αβij = interaction between treatment and phase; αrik= interaction between treatment and ecotype; βrjk = interaction between ecotype and phase; αβrijk = interaction between treatment, ecotype and phase; and εijk = error term associated with each body weight at a particular age.


The mean hatching weights observed for BN and KK were 32.27g and 34.75g respectively (Table 1). There was no significant difference between the hatch weights of the chicks under the different treatments. KK ecotype showed higher hatch weights than BN ecotype, although this was not significantly different (P<0.05).

Table 1. Mean hatch weights for the two ecotypes


Mean hatch weights (g)




























Results of the mean mature weights for the two ecotypes subjected to the different treatments are shown in Table 2. Mature body weights at week 21 were significantly different between the four treatments where treatment four (TRT 4) had the highest weight in the two ecotypes (1022 g and 1143g for KK and BN respectively).

Table 2. Mean body weights of each treatment at week 21


Mature Weight (g)





















abc Means in a row with one or more letter superscripts in common are not different (P<0.05)

Effect of feed intake, FCE and ecotype on weight gain

Table 3 presents the results of a regression analysis that was run on the data obtained from the feeding period. The model was significant (P<0.05) and it explained 92.8% of the data. Ecotype and phase were not significant to the model. Other variables (age, intake and FCE) were significant to the regression model.

Table 3. variable coefficients
















In both ecotypes, growth phase significantly affected the FCE (P< 0.05) where the second growth phase (9 -14wks) showed the highest FCE (0.432) as presented in Figure 1.

Figure 1. Feed conversion efficiency (FCE) at different growth phases.

Ecotype did not significantly (P< 0.05) affect the average daily gain. However, Bondo ecotype showed a higher mean in ADG than Kakamega ecotype (Figure2). ADG was highest between week 9 and week 15 of the growth period which coincided with growth phase 2 that had the highest FCE. Week 14 recorded the highest ADG. There was significant difference (P< 0.05) in mean ADGs between week 12, 13 and 14 as compared with the rest of the growth period.

Figure 2. Average daily gain (ADG) for Kakamega and Bondo ecotypes.


The mean body weights in table 2 observed for both ecotypes were lower than those reported by Ng’eno (2011). The lower body weight recorded for BN and KK ecotypes may be due to changes in the management regime, which shifted to a low input system of scavenging s Ng’eno(2011) used intensive production system for the same ecotypes throughout his study period. The growth performances of ecotypes studied underlined the differences between ecotypes with BN having remarkably higher growth performance levels than KK which may be attributed to differences in feed conversion efficiency (FCE) and scavenging ability. This is in agreement with results of Nge’no (2011) and Tadelle et al(2003).

The indigenous chicken ecotypes under this study were reared under the free range production system where they were occasionally supplemented using commercial feeds depending on the growth stage. They were put under high level health management corresponding to practices in commercial rearing of hybrid chicken with which farmers achieve high productivity levels. Under this management practice, the chicken showed variation in growth patterns. This is consistent with results from other countries e.g. Tanzania, Malawi and Ethiopia (Ng’eno 2011).

Despite lower hatch weight (Table 1), the BN ecotype had higher growth rates during the early stages and demonstrated better scavenging habits than KK ecotypes, which can explain the higher mean body weights observed in Table 2.This was consistent with egg weights at incubation where the variations in hatch weights follow the egg weight pattern at incubation.

A study by Tadelle et al (2003) showed a positive correlation between egg weight at incubation and body weight at hatch where 1 gram egg weight difference translates to 1.5 gram body weight difference at hatch. Magothe et al (2010) also reported a positive correlation between the egg weights and the age at lay in the parental population. Pinchasov (1991) reported that hatch weight was not affected by the age of the hen at lay but found a close correlation between egg weight and hatch weight. He reported chick to egg weight ratio at hatching to be 0.71 and independent of the hen age. One day after hatching, chick weight decreased by 1.5g. This decrease in weight was due to transitional period between the yolk as the major nutrient supplier and the time chicks start feeding on exogenous feeds (Tadelle et al 2003).The study also showed that the correlation between egg weight and chick weight decreased markedly during post-hatch growth, becoming insignificant five days after hatching.

From these studies by Tadelle et al (2003), Magothe et al (2010) and Pinchasov (1991), it is also evident that the advantage of high hatching weights attributed to large eggs diminished after hatching, which suggest that feed intake has influence on growth rate and final body weights. According to Witt et al (2004), the hatch weight is determined primarily by egg weight and secondarily by weight losses during incubation, shell and residue weight, strain, incubation time, breeder age, gender of chick and time after hatching. Chick weight normally represents 62 to 76 % of initial egg weight (Witt et al 2004). This study concludes that the correlation between egg weight and chick weight decreased with age of the chick, and at the end of 8th week, no significant correlation was recorded.

Results on feed conversion efficiency (Figure 1) showed a large variation in growth and feed utilization between the three growth phases of the two ecotypes which agrees with other previous findings from Kenya and elsewhere in the world (Ng’eno 2011, Tadelle 2001, Sonaiya et al 2000). This offers an opportunity to use phase feeding as a feeding strategy to improve on IC productivity

In both ecotypes, growth phase significantly affected the FCE (P< 0.05) where the second growth phase (9 -14wks) showed the highest FCE (0.432) which coincided with treatment TRT3 that had the highest average body weights in earlier reported results of this study. This agrees with Ng’eno (2011) whose study showed that the self-accelerating growth phase estimated by Gompertz models reached the maximum between the 10th and 14th week in all the ecotypes. However, once supplementation is stopped, the chickens were not able to maintain the accelerated growth up to maturity and therefore the group supplemented in growth phase 2 had a lower mature weight than those supplemented in growth phase 3. This is shown by results on mature weights presented in Table 2 where treatment 4 that targeted supplementation in growth phase 3 had the highest market weights.

The accelerated growth rate recorded for the chickens on TRT4 was due to compensatory growth, also referred to as catch up growth, rebound growth or rehabilitative growth (Zubair et al 2007). This occurred when the chicken were deprived food for a period of time then followed by a period of food abundance. This causes accelerated growth rate that exceeds that achieved by animals fed continuously. The chicken seem to ‘store’ growth potential where they are able to contend with fluctuations in feed supply and later release this potential once supplemented (Zubair et al 2007). This phenomenon is important economically for indigenous chicken farmers as it enables them to plan feeding schedules so that maximum use of supplements is done when the market conditions are favorable.



Badubi S S, Rakereng M and Marumo M 2006 Morphological characteristics and feed resources available for indigenous chickens in Botswana. Livestock Research for Rural Development 18:3.

Cheburet J 2010 Improve indigenous chicken. The Organic Farmer, December 2010

Food and Agriculture Organization of the United Nations (FAO) 2012 Small livestock, big impact. Accessed in April 2012.

Gondwe T N P, Ambali A J D, Chilera F C, Lwesya H and Wollny C 2005 Rural poultry biodiversity in Lilongwe and Mzuzu Agricultural Development National Statistical Office Divisions (ADD), Malawi. Malawi Journal of Science Technology. 5: 17-21.

King'ori A M, Tuitoek J K, Muiruri H K, Wachira A M and Birech E K 2007 Protein intake of growing indigenous chickens on free-range and their response to supplementation. International Journal of Poultry Science. 6:617-621.

King'ori A M, Tuitoek J K, Muiruri H K and Wachira A M 2010 Indigenous chicken production in Kenya. International Journal of Poultry Science.9: 309-316

Magothe T M, Muhuyi W B and Kahi A K 2010 Genetic parameter estimates for egg and body weights of Indigenous chicken in Kenya. World's Poultry Science Journal. 86:250-251.

Magothe T M, Okeno T O, Muhuyi W B and Kahi A K 2012 Indigenous chicken production in Kenya. World's Poultry Science Journal.68: 310-315.

Mcainsh C V, Kusina J, Madsen J and Nyoni O 2004 Traditional chicken production in Zimbabwe. World's Poultry Science Journal. 60: 233-246.

Msoffe P L M, Minga U M, Olsen J E, Yongolo M G S, Juul-Madsen H R, Gwakisa P S and Mtambo M A 2001 Phenotypes including immune-competence in scavenging local chicken ecotypes of Tanzania. Tropical Animal Health and Production. 33: 341-354.

Ng’eno K 2011 Genetic analysis of growth patterns of different ecotypes of indigenous chicken in Kenya. thesis. Pp. 4-19.

Ngeno K, Vander-Waaij E H and Kahi A K 2014 Indigenous chicken genetic resources in Kenya: their unique attributes and conservation options for improved use. World's Poultry Science Journal.Vol. 70, March 2014

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

SAS institute 1998 SAS® user’s guide: statistics version 9.1. SAS institute Cary, N.C, USA.

Sonaiya E B 2000 Family poultry and food security: research requirements in science, technology and socioeconomics. Proceedings XXI World’s Poultry Congress, Montreal, Canada, 20-24 August, 2000.

Swai E E, Karimuribo E E, Kyakaisho P F, and Mtui P 2005 Free-range village chickens on the humid coastal belt of Tanga, Tanzania: their roles, husbandry and health status Livestock Research for Rural Development 19:8.

Tadelle D and Ogle B 2001 Village poultry production systems in the Central Highlands of Ethiopia. Tropical Animal Health and Production. 33: 521-537.

Tadelle D, Kijora C and Peters K J 2003 Indigenous chicken ecotypes in Ethiopia: growth and feed utilization potentials. International Journal of Poultry Science.2:144-152.

Witt F and Schwalbach 2004 The effect of egg weight on hatchability and growth performance of New Hampshire and Rhode Island Red chicks. South African journal of Animal science. 34:62-64.

Zubair A K 2007 Compensatory growth in the broiler chicken: a review. World's Poultry Science Journal. 52:186-201

Received 9 January 2016; Accepted 3 March 2016; Published 1 April 2016

Go to top