Citation of this paper |
Surveys in the areas of Douala and Western Highlands of Cameroon were carried out in order to quantify the efficiency of broiler meat production. A total of 20 small-holder farms were included in the survey, the flock size on the farms ranging from 102 to 3060 birds.
On 8 farms commercial feed was used and on the other farms the diets were prepared using locally available feedstuffs and commercial concentrates containing animal protein, vitamins, minerals and trace elements. The chicks were obtained from local hatcheries at the age of one day. Live weight of 49-days old broilers grouped according to type of feed ranged from 1.33 to 2.06 kg. The feed conversion ratio (kg feed intake/kg weight gain) for the 49-day feeding period ranged from 1.58 to 2.44. Mortality ranged from 2.3 to 16.3%.
It is concluded that broiler production in Cameroon is
characterized by poor growth and low efficiency of feed utilization. One reason
for the poor productivity is the use of low quality feed ingredients in a
situation where the broiler industry has to expand in order to satisfy the meat
demand by an increasing human population while being in a situation of maize
shortage and prohibited use of meat meals. The objective of present and future
research therefore should be to find and test locally available, safe,
nutritionally adequate and cheap substitutes for maize and meat meal in broiler
feeding.
For many years, village poultry production has been a major
supplier of poultry meat and eggs in Cameroon. It has been credited
with more than 50% of the national production of poultry meat and eggs
(Djoukam and Teguia 1991; Tchoumboue et al 2000). Although
statistics are not available, it would appear that part of
traditional poultry production is declining along with the rapid
development, particularly around urban areas, of a modern poultry
sector specialized in the production of broilers and eggs for
consumption at national and Central Africa sub-regional levels.
Table egg production is dominated by some major industrial and
semi-industrial units, probably because of the high investment
required to enter the market. In contrast, broiler production is
mainly associated with small and medium size farms with flock sizes
varying from 50 to 20,000 birds (Joumo 1998). In the West Province
of Cameroon, housing more than 40% of the national poultry
population, 72% of the farms have flock sizes of 50 to 1000 birds
only (Djoukam and Teguia 1991). These farms are characterized by
unqualified personnel and low productivity, which could very well
be the main reason for the continuous changing identity of farmers
who leave the market as rapidly as they come in (Le Coq 2002).
Because of their fundamental contribution to the supply of chicken
meat to the national and regional markets, a strong position of the
small and medium size farms would be desired. Before any positive
action could be undertaken to support this category of farms, it is
necessary to have a clear idea of their strengths, weaknesses,
opportunities and threats. Particularly, the sector faces the
competition of imported frozen chicken meat which was to be more
than 15,000 metric tons in 2002, or the equivalent of about 10
million live broiler chickens.
The objective of this paper was to characterize broiler
production on small-holder poultry farms in order to identify the
major constraints towards its development.
To evaluate broiler production in Cameroon, surveys were
conducted in two different ecological zones, being the most
important as to the national production of broiler
meat.
The first survey was conducted in 1998 in the Douala area.
Douala is the economic capital of Cameroon situated in the coastal
plains with an altitude barely reaching 30 m above the sea level.
The climate is of an equatorial maritime type characterized by an
abundant rainfall of about 2550 mm per annum. Temperature varies
between 18.5 and 30 °C with an average temperature of 22.9
°C. Relative humidity is 60-100%, depending on the season. The
second survey took place in 2000 in the Western Highlands of
Cameroon (WHC). The WHC are situated in the Sudan-Guinean zone
characterized by an altitude that varies between 1400 and 2000 m
and a temperature of 16-27 °C. There is a single long rainy
season from mid-March to mid-November and a short dry season from
mid-November to mid-March. Annual rainfall is 1600 to 2200 mm and
relative humidity varies from 60 to more than 95%.
In the Douala area, the performance of a total of 6,034 Lohmann
chicks, belonging to 3 small farms, were studied. All birds had
been bought at one-day-old from the same local hatchery. On
arrival, flock size varied from 510 to 3060 birds. Twenty two
flocks with a total of 13,563 ISA15 Vedette and Arbor
acres chicks, belonging to 17 small-size farms, were studied in the
WHC. The chicks had been bought at one-day-old from 2 local
hatcheries and flock size varied from 102 to 1129
birds.
Poultry houses in the Douala area were all of the open-side type, with wooden or concrete-block walls and roofed with aluminium material. Wood shavings were used as litter on all cemented floors. The chicks were brooded for 4 weeks in a section of the brood-growth-finisher houses at a density of about 52 chicks/m², 36 chicks/m², 21 chicks/m² and 10 chicks/m² at age of 1, 2, 3 and 4 weeks, respectively. Heat was produced using fire wood for about 7 and 14 days during the dry and rainy season, respectively. As from the 4th week, the birds were kept at density of 10 birds/m². All birds were vaccinated against infectious bursal disease using either Bur706 ® at day-old with a booster at days 10 and 20 or using Gmboral® or Hipra Gumboro® between days 10-14 with a booster between days 23-27. Vaccination against Newcastle disease and infectious bronchitis was done at 7 or 8 days of age with a booster 2 weeks later, using H120 and Hitchner B1, respectively. Coccidiostats were used on each farm for coccidiosis prevention and/or treatment.
Two of the three farms used commercial feed from two different
manufacturers while on the third farm the feed was prepared with
locally available feed ingredients such as maize, cotton seed cake,
groundnut cake, soybean cake, wheat bran, rice bran and imported
protein/vitamins/minerals concentrate. The analysed composition of
diets used in the broiler farms of the Douala zone is given in
Table 1. For the commercial diets, the guaranteed analysis values
appearing on the feed labels are also given. On one of the farms
using commercial feed, the feed was sometimes diluted either with
maize or other cheaper ingredients such as wheat bran. Feed was
given ad libitum on one farm, whereas it was supplied in
restricted amounts on the 2 others. Water was provided ad
libitum on all three farms.
Table 1. Chemical composition (% dry matter, DM) of commercial and farm-made broiler feeds used in the Douala area |
||||||
Composition, |
Starter 1* |
Finisher 1* |
Starter 2* |
Finisher 2* |
Farm starter |
Farm Finisher |
Label values |
|
|
|
|
|
|
Crude protein, % |
21.5 |
19.0 |
23.5 |
22.1 |
- |
- |
Fat, % |
5.6 |
6.3 |
5.5 |
6.1 |
- |
- |
Crude fibre, % |
3.6 |
3.6 |
5.0 |
6.0 |
|
|
Calcium, % |
0.85 |
0.84 |
1.2 |
1.0 |
|
|
Phosphorus, % |
0.70 |
0.70 |
0.5 |
0.4 |
|
|
Analysed values |
|
|
|
|
|
|
Crude protein, % |
21.5 |
21.0 |
24.3 |
21.9 |
22.5 |
23.9 |
Crude fibre, % |
5.7 |
6.5 |
6.0 |
7.0 |
8.1 |
6.1 |
Total ash, % |
9.3 |
5.1 |
11.8 |
7.3 |
6.9 |
4.5 |
* 1and 2 correspond with different feed
manufacturers |
In the WHC area, the poultry houses were all of the open-side
type with walls made of concrete blocks, wood or clay blocks. The
floors, either cemented or not, were all covered with wood shavings
as litter. All poultry houses were roofed with aluminium
material.
The chicks were brooded in a restricted area of brood-grow
poultry houses at a density of 50-68 chicks/m2 during
the first 2 weeks and 28-40 chicks/m2 for the subsequent
two weeks. The density during the finisher phase was 10-12
birds/m2. Heat was provided by firewood with a chimney
system to expel the smoke. The birds were vaccinated at the age of
one day, 8, 12 and 18 days against infectious bursal disease,
Newcastle disease and infectious bronchitis using GumboralÒ,
Hitchner B1 or la Sota and H120 or H52 sero-type vaccines,
respectively. All vaccinations were done through drinking
water.
Six farms bought the feed from three local feed manufacturers.
The others made their own feed on the farm using locally available
ingredients (maize, cotton seed cake, groundnut cake, soybean cake,
wheat bran, corn bran and rice bran) and an imported
protein/minerals/vitamins concentrate. Feed formulas for the
farm-made feeds had been proposed by the concentrate importers.
Feed and water were freely available to the chickens. The analysed
and labelled composition of three commercial feeds used in the WHC
area is documented in Table 2. In Table 3, the analysed composition
of farm-made rations is given together with the composition as
calculated by the concentrate importers that had proposed the
ingredient compositions.
Table 2. Chemical composition (% dry matter) of commercial broiler feeds used in the Western Highlands of Cameroon |
||||||
Composition, |
Starter 1* |
Starter 2* |
Starter 3* |
Finisher 1* |
Finisher 2* |
Finisher 3 |
Label values |
|
|
|
|
|
|
Crude protein, % |
21.5 |
23.5 |
20.0 |
19.0 |
22.1 |
21.0 |
Fat, % |
5.6 |
5.5 |
- |
6.3 |
6.1 |
-- |
Crude fibre , % |
3.6 |
5.0 |
- |
3.6 |
6.0 |
- |
Calcium, % |
0.85 |
1.2 |
1.0 |
0.84 |
1.0 |
0.98 |
Phosphorus, % |
0.70 |
0.5 |
0.7 |
0.70 |
0.4 |
0.68 |
Analysed values |
|
|
|
|
|
|
Crude protein, % |
21.0 |
23.5 |
22.4 |
21.1 |
21.8 |
21.4 |
Crude fibre, % |
5.3 |
6.0 |
5.7 |
6.5 |
7.0 |
4.7 |
Total ash, % |
4.7 |
11.8 |
7.3 |
5.1 |
7.3 |
7.9 |
* 1, 2 and 3
correspond with different feed manufacturers |
|
||||
Composition, |
Starter 1* |
Starter 2* |
Finisher 1* |
Finisher 2* |
Label values |
|
|
|
|
Crude protein, % |
21.7 |
21.8 |
19.7 |
20.1 |
Fat, % |
2.9 |
2.6 |
3.1 |
2.7 |
Crude fibre , % |
4.0 |
3.9 |
4.0 |
4.2 |
Calcium, % |
0.61 |
0.41 |
0.6 |
0.41 |
Phosphorus, % |
0.40 |
0.43 |
0.4 |
0.47 |
Analysed values |
|
|
|
|
Crude protein, % |
23.6 |
24.1 |
21.1 |
21.4 |
Crude fibre, % |
4.1 |
5.3 |
6.6 |
6.5 |
Total ash, % |
7.4 |
8.0 |
6.2 |
11.1 |
1
FM = farm made |
On each farm, live weight at 49 days of age was measured to the nearest 5 g using a TRADEÒ balance. Weighing of the birds was done early in the morning. Feed consumption for each flock was calculated on the basis of data kept by the farmers. The data were used to calculate total weight gain by subtracting the average live weight of day-old chicks, and to calculate feed consumption and feed conversion ration (FCR). Samples of feed collected on the farms were analysed for crude protein, crude fibre and total ash according to the methods of the AOAC (1990).
In general, the labelled crude protein contents of the
commercial feeds were confirmed by laboratory analysis (Tables 1
and 2). The calculated protein values of the farm-made feeds were
lower than the analysed values (Table 3). Irrespective of the
origin and type of feed, the analysed crude fibre content was
consistently higher than the labelled or calculated
values.
In the Douala area, the live weight of broiler chickens at 49 days of age,
when they are supposed to be ready for the market, was 1.35 kg on all three
farms (Table 4). The birds fed on farm-made feed ate more than those fed on
commercial feed. The poorest FCR of
1.91 was recorded on the farm using farm-made feed, whereas the
other two farms had FCR values of 1.81 and 1.58, respectively.
Mortality was highest on the farm with farm-made feed, mortality
being 16%, whereas the other two farms registered 7 and 2%,
respectively.
Table 4. Broiler chicken performance on small-size farms in the Douala area |
|||
Performance |
Commercial feed 1* |
Commercial feed 2* |
Farm-made feed |
Live weight at 49 days, kg |
1.33 ± 0.7 |
1.38 ± 0.2 |
1.36 ± 0.6 |
Total mortality, % |
7.47 |
2.32 |
16.34 |
Feed consumption, kg |
2.30± 0.3 |
2.95 ± 0.2 |
3.73 ± 0.3 |
Feed conversion ratio (FCR) |
1.72 ± 0.8 |
2.13 ± 0.1 |
2.74 ± 0.8 |
*1 and 2
correspond with feed manufacturers |
The performance of broilers in the WHC area is summarized in
Table 5. Total feed consumption was on average 4.39 kg and was
similar for birds fed either commercial or farm-made feed. Feed
conversion ratio was well above 2 for all groups. Live body weight
at 49 days of age was on average 1.81 kg. Mortality was about
5%.
Table 5. Broiler chicken performance on small-size farms in the Western Highlands of Cameroon |
|||||
Performance |
Commercial feed 1* |
Commercial feed 2* |
Commercial feed 3* |
FM feed 1* |
FM feed 2* |
Live weight at 49 days, kg |
1.73 ± 0.1 |
2.06 ± 0.2 |
1.63 ± 0.2 |
1.82 ± 0.1 |
1.81 ± 0.1 |
Total mortality, % |
5.51 ± 2.4 |
6.58 ± 7.0 |
4.93 ± 1.5 |
5.91 ± 2.1 |
6.28 ± 3.4 |
Feed consumption, kg |
4.52 ± 0.4 |
4.39 ± 0.6 |
4.39 ± 0.5 |
4.34 ± 0.4 |
4.30 ± 0.4 |
Feed conversion ratio (FCR) |
2..61 ± 0.2 |
2.12 ± 0.0 |
2.69 ± 0.1 |
2.38 ± 0.2 |
2.37 ± 0.1 |
*1, 2 and 3
correspond with different feed manufacturers or protein concentrate
importers |
On all farms in the two ecological zones, total feed consumption
was lower than the 4.86 kg as recommended by the hatcheries of
ISA15 Vedette chicks. The hatcheries of Lohmann and
Arbor acres chicks indicate feed intakes of 4.74 kg from
one-day-old until 49 days of age when the birds are supposed to be
ready for marketing. The low feed intake could be related to the
type of ingredients used. The high analysed values for crude fibre
would be expected to raise feed intake due to depressed feed
utilisation (Chiengain 1981; Diambra et al 1990). The diets were
fed in powdered instead of pelleted form so that the birds may have
selected the more palatable portion of the diets, thus leading to a
lower growth rate and a subsequent low feed intake (Joly 1998).
This might explain that the observed live weight at 49 days of age
on most farms was lower than 2 kg which would not be expected for
the strains of broilers used.
The diets used on the farms could have contained proteins of
poor quality and thus have provided an insufficient amount of
essential amino acids (Buldgen et al 1996). Poorly mixed feeds
require higher safety margins of critical nutrients otherwise they
may cause sub-optimal performance (Feed International 2003). On
two farms in the Douala region, feed restriction was practiced
which could lead to inadequate supply of feed rather than to poor
feed intake. However, performance on the farm using ad
libitum was not clearly better than on the other two farms. In
general, broiler birds that are fed ad libitum may take full
advantage of their genetic potential for rapid growth. Restricted
feeding is usually envisaged earlier in the life of chicks as a
means to induce compensatory growth following starvation (Leeson
and Zubair 1997; Teguia et al 2002). On one farm in the Douala
region the commercial feed was mixed with feedstuffs. Dilution of
commercial feed with uncontrolled amounts of maize or other
ingredients will change the nutritional value of the feed. Feed
intake often is biased because the added ingredients are usually
not taken into account when calculating feed consumption. Poor
weight gain generally is a consequence of inappropriate feeding
practices and the use of low quality ingredients which are
justified by the high cost of feeding broiler birds. The efficiency
of feed utilisation was relatively poor on all the farms in the WHC
area. (Table 5). FCR values were well above 2 as opposed to the
expected values of 1.95 and 1.97 for Vedette and Arbor acres
chicks, respectively. The relatively lower FCR recorded in the
Douala area point to a better efficiency of feed utilisation, but
could relate to the poor growth of the birds and/or bias due to not
recording by the farmers of the use of maize or other ingredients.
High mortality rates in broiler production are usually related
to poor sanitation programmes and inadequate conservation of
vaccines and other medicines at the farm level (Tchoumboue et al 1992). The farm in the Douala zone that used farm-made feed had a
high mortality. High temperatures associated with high relative
humidity will create an environment in which the birds have
difficulty to control their body temperature, contributing to high
mortality.
In general, the performance of broiler production in the two ecological zones was low when compared to the expected production variables. Tchoumboue et al (1992) attributed the low production to insufficient management and lack of expertise among the farmers who do not properly understand the production process and the nutritional requirements of birds because most of them have never been formally trained. Poor sanitation and poor feeding practice are the main causes for low performance of broiler production on small and medium-size farms in Cameroon.
Broiler production on small and medium-size farms in Cameroon is
characterized by poor growth, poor efficiency of feed utilisation
and high mortality. Insufficient management skill of farmers, poor
feeding and sanitary problems seem to be the major constraints to
the activity.
The development of the poultry sector in Cameroon would benefit
from improved feeding practice. In developed countries, poultry
feeding is based on maize-soybean diets. Maize and soybean meal can
effectively make up to 95% of the diet as maize is the major energy
source while soybean meal has become the standard against which
other protein sources are compared (Leeson and Summers 1997).
Other feedstuffs used by the world poultry industry, but of minor
importance in terms of quantity, include wheat and wheat
by-products, animal protein sources, minerals and vitamins. Meals
of plant origin such as groundnut cake, canola meal are only used
in small quantities. Modern poultry production has been imported to
Africa in general, and Cameroon in particular, together with these
major feedstuffs that are considered essential for a successful
business. However, as opposed to developed countries, maize is an
important staple food in Cameroon and therefore there is
competition between humans and chickens. A major consequence is the
constant fluctuation of the maize price, with a concomitant
increase in the price of chicken feed. This is a considerable
problem because the cost of feeding represents up to 70% of the
total production costs. The evolution of the price of maize and
other major feedstuffs used in Cameroon between June 2001 and March
2002 is given in Table 6.
Table 6. Average price (FCFA/kg) of feedstuffs used in chicken feed on the Cameroon market from June 2001 to March 2002 |
||||||||||
Month/ |
June |
July |
August |
Sept. |
Oct. |
Nov. |
Dec. |
Jan |
Feb. |
March |
Maize |
100 |
120 |
140 |
150 |
100 |
100 |
120 |
150 |
170 |
186 |
CSC |
120 |
120 |
120 |
120 |
150 |
160 |
160 |
160 |
170 |
170 |
S. meal |
360 |
360 |
360 |
360 |
370 |
380 |
380 |
380 |
380 |
380 |
G. cake |
250 |
250 |
250 |
250 |
250 |
250 |
260 |
250 |
260 |
260 |
P.K.cake |
45 |
45 |
45 |
45 |
50 |
50 |
50 |
50 |
50 |
50 |
W.bran |
42 |
42 |
42 |
42 |
42 |
42 |
42 |
45 |
45 |
45 |
O. shell |
60 |
60 |
60 |
60 |
60 |
50 |
50 |
50 |
50 |
50 |
- CSC =
cotton seed cake; S. = soybean; G. = groundnut; PK.= palm kernel; W. =
wheat; O.= oyster. |
Table 6 shows that the price of maize increased by more than 80%
within a 1-year period, whereas that of soybean meal increased by
5.5% during the same period. The systematic increase in the price
of maize was due to insufficient national production. All of the
soybean meal used for animal feeding in the country is imported not
only because national soybean production is insignificant, but
there is no industry for the local production of soybean oil and
meal. Thus, importation of the two major feed ingredients, maize
and soybean meal, is the only way if poultry production is to be
based exclusively on these ingredients. This would mean that huge
amounts of foreign exchange are needed and at the same time are
depriving a poor country of resources that could otherwise be used
for development projects with positive impact on the national
economy. Furthermore, the higher price of locally produced chicken
meat would justify the import of this product from developed
countries where the cost of production is more efficiently
controlled, thus increasing the fragility of local broiler
industry.
Finding local substitutes for maize and soybean meal could help
bring down the price of poultry meat to the level that can be
afforded by of the average citizen. Imported concentrates
containing protein, vitamins and minerals represent up to 10% of
broiler rations. Their fundamental role is to provide essential
amino acids, vitamins, minerals and trace elements. The
concentrates not only are expensive, but now are subject to harsh
regulation as well. After the recent cases of "mad-cow disease" in
Europe, the import of concentrates containing meat meals has been
banned in the Central African sub-region as from December 2000. As
from February 2001, in Cameroon the use of meat meals in poultry
nutrition is prohibited even though chicken meat has not been
identified as a cause of the transmission of Creutzfeld Jacob's
disease. Meat meals have been an important source of essential
amino acids. Thus, the broiler industry in Cameroon is facing
multiple challenges. It has to expand in order to satisfy the meat
demand by an increasing human population while being in a situation
of maize shortage and prohibited use of meat meals. The objective
of present and future research therefore should be to find and test
locally available, safe, nutritionally adequate and cheap
substitutes for maize and meat meal in broiler
feeding.
* When original paper is in French, the translated title is in italic
AOAC 1990 Official method of analysis,
13th edition Association of Official Analytical
Chemists, Washington D.C.
Buldgen A, Parent R, Steyaert P et Legrand D 1996
Aviculture semi-industrielle en climat subtropical. Guide pratique.
Ed. Les Presses Agronomiques de Gembloux-Belgique, 118 p.
* Semi-industrial poultry production under sub-tropical
climate. Practical guide.
Chiengain 1981 Valeur nutritive des provendes vendues
dans les provinces du Centre-sud, Littoral, Ouest et Sud-Ouest du
Cameroun. Mémoire de fin d'études, Ecole Nationale
Supérieure Agronomique, Yaoundé, Cameroun.
* Nutritive value of manufactured feed sold in Centre-South,
Littoral, West and South-West provinces of
Cameroon.
Diambra O, Kulube K et Olaboro G 1990 Alimentation et
ressources alimentaires. Rapport de séminaire, CTA, tome 1,
17-22.
* Feeding and feed resources. Seminar
proceedings
Djoukam J et Teguia A 1991 Filière des produits
avicoles au Cameroun: Typologie des élevages avicoles
semi-intensifs dans la Province de l'Ouest. Centre Universitaire de
Dschang Cameroun., Juin 1991, 41 p.
* Poultry products pathways in Cameroon: Typology of
semi-intensive poultry farms in the Western Province of Cameroon.
University Centre of Dschang.
Feed International 2003 Achieving the proper mix. Feed International, May 2003: 26-27. Watt Publications.
Joly P 1998 Production d'œufs hors sol: quelques
conseils pratiques. Afrique Agriculture N° 259, 36-37.
* Cage table eggs production: Some practical
suggestions.
Joumo E 1998 Performances techniques de production
d'œufs de consommation des élevages de petite et moyenne
tailles dans la région de Bafoussam. Mémoire de fin
d'études, Faculté d'Agronomie et des Sciences
Agronomiques, Dschang, Cameroun.
* Production performance of small and medium holding
table egg farms in the Bafoussam area. Memoir, Faculty of Agronomy
and Agricultural Sciences, Dschang Cameroon.
Le Coq 2002 Trimestriel d'information et de formation de
la filière avicole au Cameroun. N°01, juin
2002.
* Cameroon Quarterly Poultry Information and Training
bulletin.
Leeson S and Zubair A K 1997 Nutrition of broiler chicken
around the period of compensatory growth. Poultry Science 76,
992-999.
Leeson S and Summers J D 1997Commercial Poultry
Nutrition, Second, Ed., University books, Guelph, Ontario, Canada,
355pp.
Teguia A, Awah Ndukum J and Kouomou S B 2002 Response of
broiler chickens to feed restriction during brooding under tropical
conditions. Journal of Animal and Feed Sciences 11,
127-133.
Teguia A, Manjeli Y, Tchoumboue J and Djoukam J
1992Productivité de l'aviculture intensive au Cameroun: I-
Performances techniques de production d'œufs de consommation
des petits élevages intensifs. Cameroon Bulletin of Animal
Production, 1: 1-10.
* Productivity of intensive poultry production in Cameroon:
1- Production performance of small holding table eggs production
farms.
Tchoumboue J, Teguia A, Manjeli Y et Djoukam J 1992
Productivité de l'aviculture intensive au Cameroun: II-
Performances techniques de production industrielle d'œufs de
consommation. Cameroon Bulletin of Animal Production, 1:
11-20.
* Productivity of intensive poultry production in Cameroon:
2- Production performance of industrial table eggs
production.
Tchoumboue J, Manjeli Y, Teguia A et Ewane N J 2000
Productivité et effets comparés de trois systèmes de
conduite de l'élevage sur les performances de l'aviculture
villageoise dans les hautes terres de l'Ouest Cameroun. Sciences
Agronomiques et Développement, 2: 6-14.
* Productivity and comparative effects of three
management systems on the performance of village poultry in the
Western Highlands of Cameroon.
Received 3 November 2003; Accepted 21 December 2003