Two experiments were conducted to compare the egg production of White Leghorn layers under intensive, semi-intensive and traditional household systems of rearing. In the intensive and semi-intensive systems, ten groups of 50 pullets each were randomly assigned to intensive feeding and full-day scavenging plus 120, 90, 60 and 30 g/day of balanced ration. In the rural household system, groups of 5 pullets each were distributed to 120 settlers and placed on the same 5 treatments replicated 20 times. Data on feed consumption, egg production and related parameters were collected up to the age of 50 weeks.

Intensive feeding and scavenging plus 120 g/d of supplement gave significantly higher egg production than the lower levels of supplementation. On the scavenging system egg production was linearly related with the level of supplement.

An economic analysis indicated that intensive and semi-intensive poultry production should be restricted to the vicinity of Addis Ababa due to the cost of the feed and transportation. House-hold poultry production supported about 200 eggs/year/hen obtained from scavenging and a supplementary daily ration of 90 g balanced feed and household scraps.

Poultry production is an area of animal agriculture, where human food production is relatively fast, initial capital investment is low and use can be made of available household labour. As women and children are often the primary beneficiaries of poultry production, such an intervention offers one possibility of providing animal proteins in the Ethiopian drought stricken and trypanosomiasis infected areas.

Availability, quality and cost of feed is the major constraint to poultry production in Ethiopia. Ethiopia is not self sufficient in cereal grains which form the bulk of the concentrate feeds for poultry. There are shortages of protein supplements and micro-nutrients (vitamins and minerals) which are needed for the preparation of balanced rations. Lack of fortified poultry feed makes it necessary to raise poultry on pasture in Ethiopia. The other feed problems include: lack of regional feed mills, the dependence on the supply of some feed ingredients from Addis Ababa (capital city) and an inadequate transportation system.

Under the present Ethiopian conditions the approach of semi-intensive and household poultry production systems is appealing and plausible. The objective of this study was to evaluate the egg production of White Leghorn layers kept under intensive, semi-intensive and traditional household systems of rearing in the Jimma region.

In experiment one, five hundred pullets of 16 weeks of age were randomly picked from a total of one thousand pullets. They were weighed, wing-banded, and randomly divided into ten groups of 50 pullets each. Each group was housed in separate experimental houses whose floors were raised and covered with sawdust litter. Each house was equipped with a laying box, watering and feeding troughs and provided with equal and adequate areas for scavenging.

At an age of 20 weeks, each of the ten groups of 50 layers was assigned to the five treatments in a completely randomized design with two replicates, for an experimental period of 20 weeks. The treatments were full confinement (intensive system) and all-day scavenging plus 120, 90, 60 or 30 g/day of a balanced poultry ration (18% crude protein and 10.5 MJ ME/kg purchased from Addis Ababa). Egg production, feed consumption and economic return were used as evaluation parameters.

In experiment two, groups of 5 pullets of the same age and breed were distributed to a group of 120 settlers (Jesuit Relief Service camp). Supplies of the same balanced ration (as in Experiment 1) and data collection forms were distributed to the participating farmers. A group of 5 pullets was randomly allocated to the same five treatments as in Experiment 1, replicated 20 times for a study period of 16 weeks. In addition to scavenging, food scraps were also given but the amounts could not be quantified.

Table 1: Mean values for egg production (Intensive vs scavenging plus supplements)
		Scavenging + supp.
	Inten.	120	90	60	30
Feed, g/d	Ad lib	113	90	60	30
Eggs/d	0.586	0.570	0.436	0.236	0.109
Feed/egg, g	183	198	206	254	277

The egg production performances of the layers kept under intensive and semi-intensive systems in Experiment 1 are summarized in Table 1. The general indications are that the egg production performance of the layers was linearly related to the levels of supplement offered (Figure 1).

Scavenging plus 30 g/day of supplement failed to support maintenance requirement and resulted in gradual body weight losses and death of 50% of the birds. The survivors produced 0.76 egg/week/hen. Scavenging plus 60 g/day of supplement also failed to support economically viable levels of egg production (average of 0.23 eggs/day). The groups receiving intensive feeding and scavenging plus 120 g/day of supplement were equally productive (0.57 eggs/day). These two treatment groups were significantly higher than the others (P<0.05) in mean weekly egg production.

There was no significant difference between the groups assigned to intensive feeding and scavenging plus 120 or 90 g/day of supplement, in terms of feed/egg. All these three treatment groups were found to be equally efficient with a cheaper cost of production than the others (P<0.05).

At the time of conducting these experiments, the egg price at Jimma and feed price at Addis Ababa were 0.2 birr/egg and 0.63 birr/kg, respectively. About 4 birr/tonne/km was required to move feeds from Addis Ababa to the poultry production sites. Cost-benefit data analysis showed that both intensive and semi-intensive poultry production should be restricted to within a radius of ? 80 km of Addis Ababa for reasonable profitability to be achieved.

The egg production performance of the layers in experiment two are shown in Table 2. The results were similar to those obtained at the college research station.

The results of experiment two indicate that household poultry keeping (5 layers per family) over one production year requires 75kg of supplementary feed in addition to scavenging and food scraps. A mean annual egg production of about 200 could be obtained under these conditions.

Table 2: Mean values for egg production and feed used for scavenging plus supplements on the settlers' farm
		Scavenging + supp.
	Int	120	90	60	30
Feed, g/d	Ad lib	120	90	60	30
Eggs/d	0.571	0.540	0.434	0.371	0.300

The results of this study indicate that household poultry keeping of improved breeds on a system of scavenging plus balanced supplement could be feasible provided there was a continuous supply of breeding stock and balanced feed supplements.

Acknowledgements are due to the German Development Service workers, Mr Konrad Gerards and Mr Peter Heldiman for their contributions in conduct of the experiment and statistical analysis. The Ethiopian Commission For Higher Education and Jesuit Relief Service jointly funded this research project.

(Received 1 March 1996)