Experimental validation of farmer innovations in incubation and brooding management of chickens

D R Kugonza, G Kirembe, E Tomusange-Nvule, R Lutalo and E Drani*

Department of Agricultural Production, College of Agricultural and Environmental Sciences (CAES),
Makerere University, P. O. Box 7062 Kampala–Uganda
donkugonza@agric.mak.ac.ug
* Agency for Integrated Rural Development (AFIRD), P. O. Box 27193 Kampala – Uganda
Environmental Alert (EA), P. O. Box 11259 Kampala – Uganda
* Community Development Resource Network (CDRN)/Participatory Ecological Land Use Management (PELUM),
P. O. Box 35804, Kampala – Uganda

Abstract

To evaluate the effect of using communal laying nests and innovative single occupant boxes for incubating and brooding chickens in central Uganda, this study was commissioned. A total of 36 indigenous and 18 crossbred hens were mated either to indigenous or Bovan Brown cocks. Hens either laid eggs in individual or communal nests. On turning broody, each hen was given 12 eggs to incubate and later brooded its chicks either with access to a free range at will, or was confined in a spacious incubation/brooding box over a ten week period.

The number of eggs laid in one clutch and the proportion that hatched per hen varied significantly across the treatments. Crossbred hens laying in communal nests had 10.7±0.5 eggs while those under individual nesting had 8.4±0.4 eggs; indigenous hens had 9.5±0.9 and 6.9±0.9 eggs respectively. The mean hatchability for confined hens (87.5±7.5%) was significantly higher than for free range hens (77.5±1.7%). Confined crossbred hens had better hatchability (93.9±6.1%) than those under free range (78.5±4.7%), while confined indigenous hens (72.5±9.3%) did not differ from their free ranging counterparts (79.8±9.6). Confined hens weaned most chicks (11.0) while free range indigenous hens had only 2.9. Confinement effectively cut out predation of chicks and transmission of disease pathogens. Confined birds also produced an average of 95.7 g of manure/bird/day, but for free range birds it was non quantifiable. The confinement innovation therefore improves chicken performance and should be promoted for small and medium holder farmers.

Key words: Brooding box, hatchability, indigenous, participatory technology development, restricted range

Introduction

Poultry keeping makes a substantial contribution to household food security throughout the developing world (FAO 2004). Globally, the supply of and demand for poultry products has shown a very rapid upward trajectory, with poultry now providing 28% of all meat (FAO 2011). Rural poultry production in developing countries is mainly based on small flocks of scavenging birds. As a consequence, the small scale producers are constrained by poor access to markets, goods and services; and limited skills, knowledge and appropriate technologies (FAO 2004). Between 80 to 90% of the households in rural Africa keep poultry, typically in the hands of women (Kitalyi 1998). The poultry industry in Uganda is currently composed of almost 40 million birds (UBOS 2010), majority (87.7%) of which, are indigenous chickens. A decade ago, estimates of annual egg and meat production stood at 41,000 tons of chicken meat and 1,085 million eggs (King, 2002); equivalent to a per capita consumption of 1.6 kg and 41 eggs respectively. This is lower than the then Sub-Saharan Africa average given by Aboe et al (2006). According to the 2008 Human Development Index, about 12% of women in Uganda were malnourished, 38% of children were underweight, 16% stunted and 6% wasted. The situation to date has hardly improved, and efforts need to be strengthened to provide more protein at household level, for both consumption and sale. Several Ugandan studies (e.g. Emuron et al 2010; Natukunda et al 2011) have showed that chickens hold the promise to overcome income and nutritional challenges in disadvantaged communities.

Transformation of the village poultry production system from subsistence, to a cash generating activity requires novel techniques that involve improving productivity while keeping production costs low and conserving the indigenous chicken genetic resources. Livestock keeping has always been identified as one of the income generating activities which has potential to reach the majority of the rural poor. For the very poor and often landless, poultry is the best starting point up and eventually off the poverty ladder. Kugonza et al (2008) reported that in Eastern Uganda, following the end of decades of debilitating inter-tribal cattle thefts, exchange of chickens for goats, and subsequently for cattle, was a strategy for restocking cattle in the sub-region, and revamping crop agriculture which is largely ox-plough driven.

The recommended/conventional methods of poultry keeping mainly emphasize intensive care. However, this is economically feasible only for medium scale enterprises that keep exotic/commercial broiler and layer breeds, and for households with alternative sources of income that can offset regular costs. For smallholder poultry farmers (<250 birds), there is need for their enterprise to be maintained at a minimum cost, otherwise, it would become non-viable. Economic analysis of broiler production in Kenya has showed that a flock size below 500 birds is not profitable (http://www.infonet-biovision.org). With the costs of managing chickens ever rising, especially due to feeding challenges, there is need to devise means of keeping these costs at a minimum. One such approach is to promote semi-intensive management techniques. Farmers in central Uganda are found of a number of chicken management practices which are largely a result of their own innovation. These non-conventional practices include: synchronised laying and hatching of eggs; temporary confinement of hens during incubation of eggs and brooding of chicks; and painting of free range chicks with unnatural colours such as green, blood-red etc. to protect them from predation. Lutalo et al (2010) have described such innovations and processes to validate them through joint experimentation with farmers, so as to attain participatory technology development (PTD) and its popularisation.

In this paper, the efficacy of some of these innovations in particular, the effect of using group/communal laying nests versus individual nests; and the use of a spatially restricted range during natural egg incubation and chick brooding, by use of an “incubation box” on the performance of indigenous hens and their progeny, was evaluated.

Materials and methodology

Study area

The experiment was conducted on a private farm (0° 31' 44 N, 32° 14' 55 E) owned by one of the innovators, and located in Namayumba sub-county, Wakiso district in Central Uganda at an altitude of 1168 m asl. The central region of Uganda has the highest human population and is characterised by an escalating demand for indigenous chickens and their products (Kyarisiima et al 2009). The study area was also chosen because it had a high concentration of chicken farmers, who keep indigenous and exotic chicken breeds and their crosses, under varied systems of management: intensive, semi-intensive and extensive, such that validated technologies from this study could easily be taken up by the community and up-scaled into business ventures.

Experimental birds and egg handling

Thirty six indigenous (I) hens and eighteen F1 crossbred (C) hens were mated either to Indigenous cocks (I) or Exotic cocks (E) of the brown-egg layer Bovan Brown breed, as shown in table 1. Nine hens were randomly assigned to each mating group. After mating, the hens laid eggs either in assigned communal nests or in individual nests. Eggs that were laid in communal nests were collected daily and stored in a separate room at room temperature in 30-egg trays. Trays were labelled to avoid mixing up eggs of different chicken genotypes. Candling of eggs was done to ensure that only good and fertile eggs were incubated. When the hens eventually turned broody, each hen was given 12 eggs of those originating from communal nests, to be incubated naturally. Only eggs from communal nests were used to avoid the confounding effect of age on hatchability, since the hatching quality for old eggs could have degenerated. Hens were then randomly assigned to either of two treatments. In treatment I, all hens were confined individually to a wooden “incubation box” measuring 2 m × 0.5 m × 0.2 m. In treatment II (control), broody hens had full access to free range conditions during the day.

Table 1. Experimental design
	Restricted range^a mating group			Free range^bmating group
Item	I × I	I × E	C × E	I × I	I × E	C × E
Hens/mating	3	3	3	3	3	3
Replications	3	3	3	3	3	3
Hens/mating system	9 (7)	9 (9)	9 (7)	9 (8)	9 (7)	9 (7)
Hens/restriction system	27 (23)			27 (22)
Total number of hens			54 (45)
I = Indigenous, E = Exotic, C = Crossbred Number of hens used for data analysis is in parentheses; some hens never turned broody and could not be used further; a Restricted range, confined hen during incubation and brooding b Free range, un-confined hen during incubation and brooding

The “incubation box” was made of timber boards on the sides, a wire mesh floor and an open roof covered by a movable mat made from woven sticks (Figure 1). The box was fitted with a movable partition board that left an open passage/creep window at the bottom. The partition was fitted when a hen was brooding, such that the hen occupied one half of the box with her chicks, which could walk through the window and eat a high protein chick mash in the other half of the box.

Figure 1. Incubation and brooding box (photo by William Critchley)

The treatments commenced after assembling of the birds to ensure that they had acclimatised to the study conditions. During the laying period all the birds were allowed access to the free range. During the incubation and brooding stage of study, hens were fed a 15% CP diet. Chicks were provided with a creep feed (20% CP) at free will during brooding until they were weaned. Both diets were based on maize bran, cotton seed cake, fish meal and common salt (Nacl). Water was provided ad libitum to hens and chicks.

Data collection and analysis

Egg production and feed intake of hens in each treatment category was measured per day. The number of chicks hatched per hen was recorded. Live weight of chicks, hens and cocks, morbidity and mortality of chicks, and faecal yield of confined hens were measured and recorded weekly. Hatchability was determined as the proportion of eggs incubated that hatched into chicks.The 2×2×2 factorial design was analysed using Statistical Analysis Systems (SAS, 2004) General Linear Models procedure. Means were compared, using Tukey’s Least Significant Difference. The statistical model used was:

Y_ijkl = µ + a_i + b_j + c_k + ab_ij + ac_ik + ɛ_ijkl

where Y_ijkl is a given observation; μ is the general mean common to all observations; a_i is the effect of the i^th hen breed (i = 2); b_j is the effect of the j^th cock breed (j = 2); c_k is the effect of the k^th range (k = 2); ab_ij is the effect of interaction between hen breed and cock breed; ac_ik is the effect of interaction between hen breed and range; and ɛ_ijkl is the random effects unusual to each observation.

Results and discussion

Effects of breed, communal laying and the restricted range/box

The mean number of eggs laid and hatched per hen varied significantly (P < 0.05) across the treatments (Table 2). The number of eggs produced per hen was highest in the use of communal/group laying. Crossbred hens laying communally had an average of 10.7±0.5 eggs while those under free range individual nesting had 8.4±0.4 eggs. Indigenous hens under group laying averaged 9.5±0.9 eggs while those under individual nests averaged 6.9±0.9 eggs. The results show that communal laying followed by removal of eggs, delayed development of broodiness in the hen, attributable to an inhibiting response to accumulation of eggs in the individual nest. Broodiness can be controlled by manipulation of the environment to remove stimuli that encourage nesting behaviour. Such stimuli include regular removal of eggs (Squires 2010) and depriving hens of their chicks (Eltayeb et al 2010). Nesting behaviour is stimulated by the interaction of oestrogen and progesterone, and starts just before egg laying, and the behaviour advances into broodiness, when laying is done for some weeks. Prolactin decreases the secretion of Leutenising hormone (LH), resulting into ovary regression and cessation of egg laying (Squires 2010).

Communal laying also positively influenced hatchability as eggs were collected and stored safely and hygienically, compared to when eggs are left to accumulate in the layer nests. Fertilised eggs are live and therefore, successful hatching depends on how well the eggs are taken care of from laying till setting (Ondwasy et al 2006).

The clutch size of indigenous free range chickens in our study was within the range for native chickens in the eastern Africa region, and had comparable mean values, for example, to 11.8 eggs in central Tanzania (Mwalusanya et al 2001) and 13.5 eggs in Ethiopia (Tadelle et al 2003). However, we observe that to improve the results of communal laying, serial and synchronised hatching could also be undertaken so that chicks are hatched in batches so as to ease their management. Serial hatching involves hens being used to sit on eggs continuously for two or more times by removing chicks every time they hatch and replacing them with new eggs (Ondwasy et al 2006); while with synchronized hatching, when hens that started laying within the same week reach broodiness, the first hen to reach this stage can be delayed by being given one dummy egg to sit on. This can be repeated for the 2nd and 3rd hens so that finally all the hens are set on one day, and the dummy eggs destroyed. Ondwasy and colleagues (2006) also reported that if serial hatching is coupled with synchronisation, then a farmer could hatch more chicks without using an incubator.

Table 2. Mean number of eggs laid, chicks hatched, chicks weaned and average chick weight per hen by hen breed, cock breed and range

Hen (H)

Cock (C)

Mating (M)

Range (R)

Significance

Item

I × I

I × E

C × E

Box

Free

SEM

H × R

C × R

Eggs/clutch

8.7^a

9.6^b

8.1^a

9.3^b

9.6^c

10.4^a

7.4^b

1.32

***

Chicks hatched/hen

7.1

8.3

6.1^a

8.2^b

6.1^a

8.0^b

8.5^b

9.3^a

5.6^b

1.94

***

Hatchability (%)

77.8

83.2

73.4

82.7

73.4

82.2

83.8

85.2^a

73.7^b

16.9

Chicks weaned/hen

6.2^a

8.3^b

4.8^a

8.0^b

4.8^a

7.7^b

8.4^b

9.0^a

4.7^b

1.73

***

Chick survival (%)

75.7^a

100^b

65.4^a

92.2^b

65.4^a

86.1^b

100^b

91.5^a

74.1^b

24.2

Chick weight at hatching (g)

110

117

115

115.9

111

115

120

112

119

0.02

SEM is standard error of mean
a,b,c Means in the same row and within factor with different superscripts are significantly different
*, **, ***, significantly different at p < 0.05, p < 0.01 and p < 0.001, respectively. NS, not significant at p > 0.05

The number of chicks hatched per bird was highest among hens that were confined in incubation boxes. An average of 11±1.0 chicks were hatched from each confined crossbred hen, an equivalent of 91.7% of the eggs set. However, this value did not differ significantly (P > 0.05) from that of confined indigenous hens that averaged 10.4±0.7 chicks (87% of eggs set), and yet the latter hens were lighter in body weight. A past evaluation of the incubation capacity of broody hens in Bangladesh (Azharul et al 2005) found that weight of hens (<950 g Vs >950 g) and number of eggs incubated (8 Vs 11 Vs 14 Vs 17) did not show a significant variation in hatchability. The positive contribution of confinement cages in improving hen performance has also been evaluated for the Bay of Bengal in India (Sunder et al 2005). Interestingly, confinement has other positive attributes such as terminating broodiness of nursing hens as early as four weeks after hatching, leading to resumption of egg laying and an overall improvement of eggs per hen per year of 43% (Sazzard 1993). Elsewhere, when confinement of chicks was done at two weeks, mother hens were able to come back into subsequent lay after 29.9 days in the dry season, and 32.8 days during the wet season (Lwesya et al 2004), while unconfined hens took up to 100 days to return to lay. Amin et al (2009) also reported performance improvements both for chicks and hens under separation, and in agreement with our current study, separation of chicks and supplementary feeding is highly beneficial.

More chicks (7.2±0.7) were produced from eggs incubated in boxes by indigenous hens bred to indigenous cocks, compared to 6.0±1.2 chicks from similar parents but incubated by free range hens. Egg hatchability was better among confined incubating hens across all breed matings. Crossbred hens that were confined in boxes during incubation had a hatchability of 93.9% which was higher than that of crossbred hens under free range (78.5%) but was lower than that for indigenous hens also under confined incubation (96%). These results show that the confinement of hens during incubation ensures that the hen sits on her eggs for longer periods since the brooding hen has access to feed and water within the vicinity. Our results on free ranging chickens agree with Njenga (2005) who found that free range hens in coastal Kenya had a hatchability of 78.5% on-station and 84.6% under on-farm conditions. Local hens under extensive conditions in Nigeria were reported to lay 60 to 80 eggs per year, and that this increases to 124 eggs when birds were kept in battery cages (Ibe 1990). In the current study, we also observed that indigenous hens were better at brooding than their crossbred counterparts, who lost their broodiness behaviour when chicks were still too young. Nevertheless, we observe that it is still an advantage if crossbred hens produce more eggs and also incubate. It is largely expected that with an increase in the proportion of exotic genes, for instance through upgrading, broodiness is likely to be lost further. FAO (2010) reported that domestic hens bred for high egg production have more or less lost their ability to become broody, because many generations of selection for higher egg production have favoured genes that hinder the onset of the broody period.

The number of chicks weaned per hen in our study followed a dramatic trend (Table 2). Chicks fathered by Bovan cocks had better survival that those from indigenous cocks, irrespective of brooding management. This could be attributed to high chick weights at hatching that are expected for chicks with some exotic genes, nevertheless, an earlier study by Hartmann et al (2002) showed that no significant relationship exists between chick weight and post hatch survival. Chicks weaned per hen were highest among crossbred hens kept in boxes (11.0) and lowest for indigenous hens under free range brooding (2.9) for similar reasons cited above. Chick survival was maximum from hatching to four weeks of age among crossbred hens kept in boxes, and was significantly high (96.4%) for indigenous hens under the same system. Results seem to show that chicks from crossbred hens under free range had maximum survival though this is not true since the chicks had to be brooded indoors, separate from their mothers, since the hens started abandoning them at one week of age. Artificial brooding is hence responsible for the observed “maximum” survival. We observe that this is the main flaw in using crossbred hens to brood chicks under free range conditions, and our study shows this is not possible. On the other hand, it is possible to utilise indigenous foster hens to brood the chicks from the crossbred hens if artificial brooding is not possible. Ondwasy et al (2006) reported that a foster hen can brood up to 65 chicks of different ages at the same time, and that when it gets cold, the youngest chicks are the first to go under the hen and the oldest will come later around the hen.

Findings of this study also show that the box effectively cuts out predation of chicks and transmission of disease pathogens. Chick mortality under free range brooding was a result of preying by wild birds and mammals; and trampling by humans and cattle. In a related study, Lwesya et al (2004) also found that confinement of chicks resulted in better survival, and highlighted that this confinement had to last beyond four weeks, otherwise the response does not differ from chicks not confined at all. In general, they also found that confinement could ensure 100% chick survival while in the control, only 40% survived. According to these authors, when chicks were released for scavenging when they were larger and heavier (than the control group), they were able to compete well for the scavenging feed resource base, while being clever and old enough to run away from predators. Gondwe and Wollny (2007) reported that in growing and adult chickens, predation contributes over 20% of losses. While this could imply that confinement is the best mode of management, the economics of enclosing and supplementing chickens vis-à-vis the potential value of the increased annual productivity should dictate the choice of management options (Lwesya et al 2004).

Other than producing many eggs which also have a better hatchability, crossbred hens do not seem to have much advantage over their pure indigenous counterparts under the free range. Considering an earlier study on a crossbreeding programme in Uganda, Sørensen and Ssewannyana (2003) poignantly showed that as the crossbred birds grew beyond three months, superiority in growth rate diminished gradually and mortality rose significantly.

Variation in hen weight, supplemental feed intake and manure yield

Weight of hens declined with time across the laying cycle and breeds (Table 3). This has also been observed elsewhere, for instance Azharul et al. (2005) reported weight losses of up to 12% during incubation. However, results of our study show that confining the hen in a box during incubation reduces rate of weight loss (Figure 2). In our preliminary work, it was observed that during the incubation period (week 4 to week 7), offering supplement feed to free range hens led to better live weights than for hens under full free range scavenging, though the hens under the latter system seemed to recover the mothering stress faster after hatching of its clutch. It should still be noted however, that incubating hens exhibit a depressed appetite (Eltayeb et al 2010) due to hormone prolactin which tends to promote broodiness at the same time.

Table 3. Effect of hen breed and range on weekly live weight of hens and weight of manure produced per hen
	Live hen weight (kg)							Manure yield/bird (g)
	Hen breed (H)		Range (R)		Significance			Range (R)
Week	I	C	Box	Free	H	R	H x R	Box	Free
1	1.69^a	1.80^b	1.85^a	1.64^b	*	***	NS	449.1	-
2	1.72	1.76	1.80^a	1.68^b	NS	**	NS	-	-
3	1.69^a	1.81^b	1.84^a	1.65^b	*	***	NS	608.4	-
4	1.69^a	1.75^b	1.79^a	1.64^b	***	**	***	-	-
5	1.62^a	1.73^b	1.75^a	1.60^b	*	**	NS	724.3	-
6	1.54	1.59	1.62^a	1.52^b	NS	**	*	-	-
7	1.54	1.59	1.63^a	1.50^b	NS	**	NS	708.2	-
8	1.54	1.62	1.65^a	1.51^b	NS	**	**	721.1	-
9	1.53^a	1.64^b	1.69^a	1.49^b	*	***	NS	704.2	-
10	1.55^a	1.65^b	1.70^a	1.50^b	*	**	NS	708.2	-
I = Indigenous, C = Crossbred a,b Means in the same row and within factor with different superscripts are significantly different , , **, significantly different at p < 0.05, p < 0.01 and p < 0.001, respectively. NS, not significant at p > 0.05

The mean supplemental feed intake per bird ranged between 498 g in the first week of confinement and 941 g during the second week of brooding the chicks, with an overall daily mean consumption across all hen groups of 108 g per bird. The mean collectable yield of dry faecal matter (manure) per bird across the entire study period was 95.7 g per bird per day, which projects to 34.9 kg of manure per bird per annum. Under smallholder peri-urban farming, poultry manure is very useful for vegetable production, and in many cases is a limiting factor. For instance, Mugerwa et al (2011) found that cabbage production by smallholder crop-livestock farmers in central Uganda was not profitable when poultry manure was applied at a rate of >4 tonnes per ha, primarily due to the high cost of this manure type. Therefore, the box innovation evaluated in this study can be incorporated into back-yard farming and in the process ensure regular poultry manure harvesting. This manure yield cannot be realised under the conventional pure free range scavenging system of poultry management, as the birds drop it on the wider range, and not necessarily on the target crop if at all.

Conclusion

Recommendations

Acknowledgements

We are grateful to the anonymous reviewer whose comments have enriched this manuscript. This work was commissioned and funded by PROLINNOVA-Uganda, an effort that promotes local innovation in agriculture and natural resource management. We are also grateful to R. Namussu and P. Mulindwa for participating in this study.

References

Aboe P A T, Boa-Amponsem K, Okantah S A, Dorward P T and Bryant M J 2006 Free-range village chickens on the Accra Plains, Ghana: Their contribution to households. Tropical Animal Health and Production 38: 223–234.

Amin M J R, Howlider M A R and Ali M A 2009 Effects of chick separation and feeding on the performance of hens and chicks. The Bangladesh Veterinarian 26(1): 13–16.

Azharul I M, Ranvig H and Howlider M A R 2005 Incubating capacity of broody hens and chick performance in Bangladesh. Livestock Research for Rural Development, Vol. 17, Art. #19. Retrieved from http://www.lrrd.org/lrrd17/2/azha17019.htm

Eltayeb N M, Wani C E and Yousif I A 2010 Assessment of broodiness and its influence on production performance and plasma prolactin level in native chicken of the Sudan. Asian Journal of Poultry Science 4(1): 1–6.

Emuron N, Magala H, Kyazze F B, Kugonza D R and Kyarisiima C C 2010 Factors influencing the trade of local chickens in Kampala city markets. Livestock Research for Rural Development, Vol. 22, Art. #076. Retrieved from www.lrrd.org/lrrd22/4/emur22076.htm

FAO (Food and Agriculture Organisation) 2011 World livestock 2011 – Livestock in food security. Rome, FAO. 115 pp.

FAO (Food and Agriculture Organisation) 2010 Chicken genetic resources used in smallholder production systems and opportunities for their development, by P. Sørensen. FAO Smallholder Poultry Production Paper No. 5. Rome. 53 pp.

FAO (Food and Agriculture Organisation) 2004 Small-scale poultry production – technical guide. FAO Animal Production and Health manual, Rome, FAO. 119 pp.

Gondwe T N and Wollny C B A 2007 Local chicken production system in Malawi: Household flock structure, dynamics, management and health. Tropical Animal Health and Production 39: 103–113.

Hartmann C, Strandberg E, Rydhmer L and Johansson K 2002 Genetic relations between reproduction, chick weight and maternal egg composition in a White Leghorn line. Acta Agriculturae Scandinavica, Section A – Animal Science 52(2): 91–101.

Ibe S N 1990 Utilizing local poultry gene resources in Nigeria. In the Proceedings of the 4th World Congress on Genetics Applied to Livestock Production, Edinburgh, UK. Vol. 16: 51–53.

King A 2002 Livestock and livestock products. In: Joint donor agencies study on the performance of and growth prospects for strategic exports in Uganda. EU Commission, Uganda. 53 pp.

Kitalyi A J 1998 Village Chicken Production Systems in Rural Africa. Household Food Security and Gender Issues, FAO Animal Production and Health Paper 142; (Food and Agriculture Organization of the United Nations, Rome).

Kugonza D R, Kyarisiima C C and Iisa A 2008 Indigenous chicken flocks of Eastern Uganda: I. Productivity, management and strategies for better performance. Livestock Research for Rural Development, Vol. 20, Art. #137. Retrieved from www.lrrd.org/lrrd20/9/kugo20137.htm

Kyarisiima C C, Kugonza D R and Magala H 2009 Analysis of production and the marketing chain of local chickens in central Uganda. Final research project report. Network of Ugandan Researchers and Research Users (NURRU), Kampala, Uganda. 50 pp.

Lutalo R, Tomusange N, Kirembe G, Kugonza D R, Lutalo S and Critchley W 2010 Trying out joint experimentation in poultry farming in Uganda: an experiment in itself. In: Wettasinha C and Waters-Bayer A (eds.) Farmer-led joint research: Experiences of PROLINNOVA partners. A booklet in the series on Promoting Local Innovation (PROLINNOVA). Silang, Cavite, Philippines: IIRR / Leusden: PROLINNOVA International Secretariat, ETC EcoCulture, pages 64–69.

Lwesya H, Phoya R K D, Safalaoh A C L and Gondwe T N P 2004 Rearing chicks in enclosures under village conditions: effect on chick growth and reproductive performance of mother hens. Livestock Research for Rural Development, Vol. 16, Art. #89. Retrieved from http://www.lrrd.org/lrrd16/11/wesr16089.htm.

Mugerwa S, Kabirizi J M, Kigongo J and Zziwa E 2011 A cost-benefit analysis for utilisation of poultry manure in cabbage production among smallholder crop-livestock farmers. International Journal of Agronomy and Agricultural Research 1(2): 14–19.

Mwalusanya N A, Katule A M, Mutayoba S K, Mtambo M M A, Olsen J E and Minga U M 2001 Productivity of local chickens under village management conditions. Tropical Animal Health and Production 34(12): 405–416.

Natukunda K, Kugonza D R and Kyarisiima C C 2011 Indigenous chickens of the Kamuli Plains in Uganda: I. Production system and flock dynamics. Livestock Research for Rural Development, Vol. 23, Art. #220. Retrieved from http://www.lrrd.org/lrrd23/10/natu23220.htm.

Njenga S K 2005 Productivity and socio-cultural aspects of local poultry phenotypes in coastal Kenya. The Royal Veterinary and Agricultural University, Copenhagen. (PhD thesis).

Ondwasy H, Wesonga H and Okitoi L 2006 Indigenous chicken production manual. Kenya Agricultural Research Institute, KARI Technical Note No. 18, Nairobi, Kenya. 13 pp.

SAS (Statistical Analysis Systems) 2004 SAS OnlineDoc Version 9.1.3. English edition, SAS Institute, Inc., Cary, North Carolina, USA.

Sazzad M H 1993 Manipulation of the broody period to increase egg production of indigenous hens under rural conditions in Bangladesh. Livestock Research for Rural Development Vol. 5, Art. #2. Retrieved from http://www.lrrd.org/lrrd5/2/bangla2.htm

Sørensen P and Ssewannyana E 2003 Progress in SAARI chicken breeding project – analyses of growth capacity. In Proceedings of the Livestock Systems Research Programme Annual Scientific Workshop, held March 2003, Kampala, Uganda, DANIDA’s Agricultural Sector Research Programme (ASPS) and National Agricultural Research Organisation (NARO) Uganda. pp. 172–178.

Sunder J, Chatterjee R N, Rai R B, Kundu A, Senani S, Singh A K and Jeyakumar S 2005 Production performance of indigenous and crossbred poultry germplasm of Andaman and Nicobar Island. Indian Journal of Animal Science 75: 1326–1328.

Tadelle D, Million T, Alemu Y and Peters K J 2003 Village chicken production systems in Ethiopia: 1. Flock characteristics and performance. Livestock Research for Rural Development (15)1. Retrieved from http://www.lrrd.org/lrrd15/1/tadea151.htm

UBOS (Uganda Bureau of Statistics) 2010 Statistical Abstracts. 250 pp. Retrieved from http://www.ubos.org