Livestock Research for Rural Development 19 (8) 2007 | Guide for preparation of papers | LRRD News | Citation of this paper |
A cross-sectional survey investigating roles, husbandry and health status of free-range village chickens (FRVC) was carried out in three administrative wards of Muheza district (now Mkinga), Tanga region during the period of June 2006. Responses were provided by 27 men and 16 women. The mean (SD) household flock size was 13.2 (11.6) and the median (range) was 9 (1-63). Half of the respondents claimed to keep chickens for meat and eggs, 43% claiming to keep them for household cash security and 2% for socio-religious functions at homes.
Fifty one percent of chickens were reported to be housed in a specific chicken banda made of local material. Forty eight percent (n=21) of the visited households were keeping chicken at their own residential houses and in some cases in the kitchen. Chickens were owned by individual members of the household, but women and children were the predominant providers of care for chickens. Feeding system was reported to be mainly scavenging (100%) although supplementary feeds were also provided (72%) on irregular basis.
Perceived annual mortality was estimated at 90% and 49% for chicks and adult birds respectively. Approximately 67.4% and 20.9% of respondents respectively, named Newcastle disease and Fowl pox as the most common diseases affecting FRVC. It was concluded that the low productivity of chickens was partly due to the prevailing poor management practices, in particular the lack of proper health care, poor nutrition and housing.
There is an urgent need for the government to support free-ranging poultry farmers by providing subsidized vaccinations and technical support in order to develop and stimulate economic development in impoverished rural areas of Tanga region.
Key words: diseases, free-range village chickens, husbandry, predators, roles, Tanga, Tanzania
Projected livestock census of 2002 indicates that Tanzania has a total of 47 million poultry. Of these, 27 million are free-range village chickens (FRVC) (Gallus gallus domesticus), kept mainly in the rural areas by women and children, with 80% of livestock-keeping households depending on poultry for their livelihood (Anon 2002a). Of the traditional raised birds kept in Tanzania, majorities are chicken (94.1%) followed by ducks and gees (5.3%), guinea fowl (0.4%) and turkey (0.2%) (Melewas 1989).
Rural chickens are characterized by low production coefficients (Minga et al 1987; 1999; Kitalyi 1998). Research work on local chicken carried out in Tanzania and elsewhere in Africa shows that the number of clutch per year is 2-4; egg per clutch is 15; egg weight is 44-49g; hatchability is 78% and chick mortality is 32.6% (Guèye 1998; Kitalyi 1998; Mwalusanya et al 2002). The low productivity is further amplified by diseases, poor genetical potential, poor bird management practices, predators and low level of literacy among farmers (Minga et al 1987, 1999; Katule 1988; Msoffe et al 2002; Salum et al 2002; Conroy et al 2005). Advantages of the traditional system include availability of free feed resources in the surrounding environment and kitchen leftovers, use of local breeds that are adapted to their environment and preserved ability to incubate and brood naturally (Magwisha et al 1997; Pedersen et al 2002).
To date, there are no detailed studies conducted targeting comprehensive description of the flock characteristics and associated husbandry, management practices of the village production system in the coastal belt zone of Tanga region in Tanzania. Understanding the roles and function of local chicken as well as production constraints is of considerable relevance in envisaging future research and development directions and strategies.
Participatory Rural Appraisal (PRA) surveys (Anonymous 2002b) highlighted and acknowledged the importance of rural village chicken as the source of protein and household income in the surveyed villages. The survey further revealed the problem of ectoparasitosis due to soft ticks, locally known as "Paazi" to be the main vector constraint. The later was identified and verified to be Argas persicus by the fore runner study survey, which did elucidate further on the biology and role of the ticks on birds (Swai et al 2007).
In an attempt to determine the actual husbandry and bird disease
prevalent in the area as well as to evaluate the existing
socio-economic role they contribute to the household, two villages
in Mkinga and one village in Maramba divisions, Muheza district
(now Mkinga) were selected for study. This paper reports the status
of bird management husbandry practiced and the disease prevalent in
the three villages.
This survey was conducted in two Divisions of Mkinga and Maramba in Muheza district (currently located in the new district of Mkinga) in Tanga region during June 2006. Study divisions are located between Latitude 4 and 6 South of Equator and Longitude 36 and 38 East of Greenwich. Three administrative wards, namely Moa, Manza and Gombero comprising of Mahandakini, Gezani and Vunde Manyinyi villages respectively, participated in the study. Mahandakini village is located about 55 km north of Tanga city while Gezani village is located 30 km north of Tanga city. On the other hand, Vunde Manyinyi is located 29 km North East of Tanga city. According to District Agriculture office; all three villages are classified and categorized to be located in the coastal belt zone which is characterized by low altitude, wet, humid with two distinct rainy seasons namely: 'Masika' (March-May) and 'Vuli' (September-October) interspersed by short rains known as 'Mchoo'. Details of study sites are described elsewhere by Swai et al (2007). The three study wards represent 2.5, 2.0 and 4.1% of all human population of Muheza district and the estimated household population size is 4.5(URT 2002, Table 1).
Table 1. Moa, Manza and Gombero wards: human population at glance |
|||
Variable |
Ward |
||
Moa |
Manza |
Gombero |
|
Total population, n |
7,044 |
5,671 |
11,443 |
Proportion females, % |
52.8 |
52.6 |
48.6 |
Proportion males, % |
47.2 |
47.4 |
51.4 |
Percent of Muheza1 district female population, % |
2.6 |
2.1 |
3.9 |
Percent of Muheza1 district male population, % |
2.3 |
1.9 |
4.2 |
Percent of Muheza1 district population, % |
2.5 |
2.0 |
4.1 |
Source: own calculation based on 2002 census (URT 2002) 1Percent of Muheza district area and population refers to the former Muheza district before it was divided into Muheza and Mkinga districts in July 2005 |
Households for the study were randomly selected based on the past experience of chicken keeping, possession of chicken(s) and willingness to participate in the study. Therefore households that participated in the current study included 11 in Vunde Manyinyi, 20 in Gezani and 12 in Mahandakini village.
Secondary data related to chicken production were collected through consultative and participatory procedures from various sources including district agriculture and livestock (department) office. Literature on indigenous chicken production system, constraints and opportunities were sought and reviewed. Primary data collection was carried out through personal interview of household members in selected villages, which was coupled with checklist and farm inspection to validate interviewee's information. The data collected include: socio-demography of chicken keepers, flock size, composition, ownership, type of housing, feeds and feeding practices, diseases and perceived flock mortalities, use pattern of chicken and animal health service provision. Flocks were categorized as follows: Chicks (aged between 1- 3 months), middle aged (>3-9 months) and adult (aged > 9 months). In addition, focus group discussion with some farmers was used to collect information on importance of diseases using matrix-ranking technique. Data collection was preceded by a short introduction of objectives and purpose of the investigation to the local extension officer and villagers (Plate 1).
|
|
A database was developed to store quantitative data from this
cross sectional study using Epi Info version 6.04d software (CDC,
Atlanta 1996). The same programme was used to compute descriptive
statistics of variables collected during the study. Comparison of
proportions of independent variables was done using Chi-squared
(c2) test while Analysis of Variance (ANOVA) was used to
test existence of significant difference between means of the
continuous variables. The Bartlett's test for inequality of
population variance was used to assess homogeneity of variances and
a small p-value (p<0.05) was used a basis for adopting the
Kruskal-Wallis test of statistical significance of the two groups
compared instead of ANOVA. A critical probability of 0.05 was
adopted throughout as a cut-off point for statistical significance
between groups compared.
Overall, indigenous chicken farmers interviewed were mainly Digo (74.4 %), Zigua (9.3%), Nyamwezi (4.7%) and Sambaa (4.7%). Others were Makonde (2.3%), Bondei (2.3%) and Shirazi (2.3%). These are the main ethnic groups residing in the study area (Table 2).
Table 2. Socio-economic characteristics of the surveyed farm by ward |
||||
Attribute |
Ward n, % |
Over all n, % |
||
Moa |
Manza |
Gombero |
||
Level of education |
|
|
|
|
No formal school |
4(36.4) |
7(35) |
4(33.3) |
15(34.9) |
Primary education |
7(63.6) |
12(60) |
6(50) |
25(58.1) |
Secondary school |
0(0) |
1(5) |
2(16.7) |
3(6.9) |
Total |
11 (100) |
20 (100) |
12 (100) |
43 (100.0) |
Ethnic group |
|
|
|
|
Bondei |
0(0) |
0(0) |
1(8.3) |
1(2.3) |
Makonde |
1(9.1) |
0(0) |
0(0) |
1(2.3) |
Digo |
6(54.5) |
16(80) |
10(83.3) |
32(74.4) |
Nyamwezi |
1(9.1) |
1(5) |
0(0) |
2(4.7) |
Sambaa |
2(18.2) |
0(0) |
0(0) |
2(4.7) |
Zigua |
1(9.1) |
2(10) |
1(8.3) |
4(9.3) |
Shirazi |
0(0) |
1(5) |
0(0) |
1(2.3) |
Head household |
|
|
|
|
Females |
5(45.5) |
7(35) |
4(33.3) |
16(37.2) |
Males |
6(54.5) |
13(65) |
8(66.7) |
27(62.8) |
Out of the 43 households that participated in the study, 27(62.8%) were headed by males and only 16 (37.2%) were headed by females. Distribution of the survey respondents by level of education is summarized in Table 2. Majority of the respondents were primary school leavers (58.1%), fifteen (34.9%) had no formal education and three (6.9%) were secondary school leavers. There was no significant difference in proportion of respondents with respect to education levels in the three wards surveyed (P>0.05). Mean age of the interviewed farmers was 49.5±16.4 (mean±SD) years and the age ranged between 25 and 66 years.
The average number of local chicken owned per household was 13.2±11.6 (mean±standard deviation) which ranged between 1 and 63 chickens per household. In all three villages visited, the average number of chicks (4.9± 1.19) was significantly higher than middle aged (3.27±0.75) and adult chickens (4.04±0.57) (P<0.05). The number of females birds was significantly higher than male in each household interviewed (P<0.05) (Table 3).
Table 3. The average number of chicken by age category and sex |
|||
Age category |
Number of birds (mean±SD) |
Overall |
|
Females |
Males |
||
Chicks |
3.20±5.10 |
1.67±3.1 |
4.90±7.8a |
Middle aged |
2.40±3.50 |
1.27±2.24 |
3.27±4.6b |
Adult |
3.11±2.73 |
0.90±1.60 |
4.04±3.74b |
ab Least square means with the different superscript within a column are significantly different (P < 0.05) |
The median and mode of total chicken per household was 9 and 20, respectively. Flock size was not significantly (P> 0.05) influenced by the sex of the head of the household. The flock size in this study (13.2±11.6) was smaller that what has been reported in Morogoro region in Tanzania and in most of the Southern African countries where flock size ranged from 10 to 60 (Kelly et al 1994; Mwalusanya et al 2002; Mapiye and Sibanda 2005; Mushi et al 2006).
All farmers reported to provide housing to their chickens reflecting the awareness on importance of housing. However, most of bird houses were of poor quality and hence there is need to educate them to build proper housing structures so that they realize more benefits. In 51.2 %( n=22) of the households, specific chicken 'banda' (or small hen house) made of local material was observed. Forty eight percent (n=21) of the visited households were keeping chicken at their own residential houses. In some cases the chicken were housed in the kitchen (Plate 2) near family house.
|
|
Keeping birds at residential houses was observed to provide
assured security from both thieves and predators than chicken banda
constructed outside/ adjacent to residential houses. Proper housing
must not only provide an environment that moderates environmental
impact but must provide adequate ventilation for birds to lay eggs
in nest boxes, as well as to feed and sleep in comfort and security
(Smith 1992). Lack of adequate housing can partly explain chicken
mortalities and thus good housing is therefore, a prerequisite for
any viable and sustainable chicken project.
Although 62.8% of the households were male-headed, women owned the bulk of the chickens, with an overall ownership of 66% followed by children (15%) and husband (12%). Scavenging was the major feeding system (100%). Provision of supplementary feeds was reported to be practiced by 31 (72%) of the farmers interviewed and it was reported to be irregular. These results are in agreement with reports from other developing countries where scavenging was noted to be the main feeding system for village chicken (Mwalusanya et al 2002; Pedersen et al 2002). The level of supplementary feeding did not vary between the three administrative wards surveyed (P>0.05). The most common feeds offered as supplements were mainly cereals. These were in the order of importance: sorghum (n=13, 41.9%), coconut waste (n=11, 35.5%), and maize bran (n=7, 16.2%). Watering was not reported to be offered by all of the households visited and chickens were reported to search for water on themselves during daytime.
Respondents indicated variety of use pattern of their chicken. Self- or home (chicken/eggs) consumption was the first (n =24, 55%) chicken use mentioned by farmers. Village chickens were mostly slaughtered for guests, but also sometimes during festivals like Christmas, New Year and Idd el Fitri. Other social-cultural use pattern includes giving relatives and friends chickens as gifts or as tokens of appreciation for services rendered. Sale for cash earning only represented (n=9, 23 %) of the total use. Chickens were also sold to pay for school fees, medical costs and paying village taxes. Sale price was on average Tshs 1,000-2,000/= (middle aged), Tshs 2,000-3,000/= (adult females) and Tshs 2,500-3,500/= (adult males) (Exchange rate was approximately 1US$~ 1320 Tshs). In this way and in accordance with Smith (1992) and URT (2005) they largely contribute to protein malnutrition and poverty alleviation, respectively. Other use patterns of local chicken are shown in Figure 1.
|
|
Live birds and eggs are usually sold in local markets and occasionally to middlemen for retail in the medium to large town like Tanga. According to farm households, the largest off-take rates from the flock occur particularly during holidays and festivals and during the onset of disease outbreaks. The latter is meant to prevent or minimize expected financial losses from high morbidity and mortality.
During interviewing session, farmers were asked to
mention different chicken disease prevalent in their area and rank
them in order of priority. Out of 43 respondents interviewed, 40
(97%) reported different disease conditions occurring in Mkinga and
Maramba Divisions. Overall, the most common diseases mentioned were
Newcastle (n=29, 67.4%) and Fowl pox (n=9, 20.9%) and the two
conditions were ranked as the most important affecting their
chicken (Table 4).
Table 4. Major disease conditions mentioned by farmers (n=43) to occur in Mkinga and Maramba divisions |
||
Variable |
Disease type |
|
Newcastle, n (%) |
Fowl pox, n (%) |
|
No. responded |
29, (67.4) |
9, (20.9) |
Average number of times mentioned (mean±StdDev) |
1.05±0.16 |
1.01±0.15 |
Range |
1-2 |
1-2 |
Other conditions reported to occur less frequently were Paralysis (5.6%), Helminthoses (3.1%), scaly leg (2.3%) caused by Cnemidocoptes mutans or burrowing mite (Robin, 1860) and Fowl Typhoid (1.5%) (Figure 2).
|
|
High disease levels were probably due to exposure of chickens to the natural environment, interaction of different entities, within and among flocks such as flock contacts during scavenging, uncontrolled introduction of new stock, contact through exchange or sale of live chickens or movement between households and villages.
The respondents were also asked to mention different chicken ecto-parasites often seen or found in their chicken and also were asked to rank them in order of priority. Out of 43 respondents interviewed, 43 (100%), 42(97.6%) and 40 (93%) reported to have seen 'Stick tight fleas' or 'Echidnophaga gallinacea (Westwood, 1875)', 'Paazi' or 'fowl tick' or 'Argas persicus (Oken, 1881)' and 'Utitiri' or 'chicken mite' or 'Ornithonyssus bursa (Berlese, 1888)' in their flocks respectively. The three-ectoparasite species were ranked as the most important affecting their chicken (Table 5).
Table 5. Ectoparasites mentioned to occur in Mkinga and Maramba divisions by farmers (n=43) |
|||
Variable |
Ecto-parasite type |
||
Stick tight Fleas, n (%) |
Fowl tick, n (%) |
Chicken mite, n (%) |
|
No. responded |
43, (100) |
42, (97.6) |
40, (93) |
Average number of times
|
1.83±0.06 |
1.81±0.07 |
1.02±0.08 |
Range |
1-3 |
1-3 |
1-3 |
Ectoparasites in the course of feeding on birds cause irritation, uneasiness when roosting, suck blood and cause anemia. Specifically, fowl tick (Walker et al 2003) causes paralysis and they are potential vector of poultry diseases including spirochetosis and aegyptianellosis . Other negative impact includes reduced weight gain, lowered egg production, toxemia and depression (Permin et al 2000).
Two distinct classes of predators were identified. These were aerial or two legged and nocturnal or four-foot predators. Overall, the first three aerial or two legged predators of priority to farmers in the surveyed villages in descending orders include 'Kipanga' or 'Falcon spp/ or kite or milvus migrans' (1st), 'Mwewe' or 'Sparrow hawk' (2nd) and 'Kunguru' or 'Crow' (3rd) and there was no significant difference in ranking between sub-villages and villages (P>0.05). 'Kipanga' was acknowledged by almost all respondents for her ability to grab young chicken (chicks). There was a striking difference when ranking was considered to land surface or four-foot or nocturnal predators. 'Kicheche' or 'mangoose' was ranked first, followed by 'Njuzi'. Predation was mainly due to the effect of 'Kicheche' (69.3% of the predation cases) followed by 'Njuzi' (30.7% of the cases). ('Kicheche' is a stinking little animal like a cat whereas 'Njuzi' is a small animal in the cat family that preys on chickens) (Table 6).
Table 6. Ranking of chicken predators by respondents in the surveyed villages |
||||
Type of predator |
Predator (Common name) |
Average score assigned by respondents |
||
Mean |
(± SE) |
Range |
||
2- legged |
Kipanga (Eagle/ Falcon, Kite) |
2.07 |
± 0.12 |
1-4 |
Mwewe (Sparrow hawk) |
1.52 |
± 0.12 |
1-3 |
|
Kunguru (Indian crow) |
1.45 |
± 0.10 |
1-3 |
|
4- legged
|
Kicheche (Skank) |
2.71 |
± 0.16 |
1-4 |
Njuzi |
2.53 |
± 0.45 |
1-4 |
|
Snake* |
1.02 |
± 0.05 |
1-3 |
|
Stray dogs |
NS |
NS |
NS |
|
Feral cats |
NS |
NS |
NS |
|
NS =not significant, * Snake is not a four foot species |
Other predators that were reported to occur in less frequency included snake and domestic carnivores (stray dogs and feral cats). These class of predators posed danger mainly during the night. Properly constructed poultry houses that can be used for nesting chickens at night would limit losses due to them. Predators are known to cause considerable losses of chicken in many rural setting in Tanzania (Msoffe et al 2002). Similar observation was obtained in Ethiopia that predators were responsible for over 70% of chick losses (Negesse 1993). Though no accurate result can be cited, there is belief that the color of chicks might be one of the attractive factor for birds of prey. White feathered chicks might be much more affected. Any chicken breeding programme targeted for rural setting should consider this important component. Therefore before any kind of colored chicks are distributed in the rural areas where chicks are raised in the open, due consideration must be given to prevalence of predators.
Perceived mortalities in chicks and adult group flock were estimated to be 90% and 49% respectively. Diseases and parasites (ecto) contributed markedly to high flock mortalities. Mortality was highest in the chick flock group. High mortality rates in chick flock group could have been due to a low physical defensive mechanism, weak and under-developed immune system or lack of knowledge by the farmers on the use of modern veterinary medicine. Mwalusanya et al (2002) recorded a survival rate of 59.7% in village chick (1-10 weeks old) flock in Morogoro.
Most owners (72.1%) complained of a lack of technical and veterinary advice. Few group of farmers (n=12, 27.9%) who had access to veterinary service responded differently in times of disease occurrence. In part, lack of intervention response by the farmers was attributed to lack of cash to purchase veterinary medicine and shortage of veterinary and extension services especially of now when veterinary services are supposed to be delivered by private practitioners. Preventive measures offered were vaccination against Newcastle using commercial (La Sota® Hipra Laborotorios, SA, Spain) and local (thermostable, I-2, ADRI® Temeke, Dar-es-Salaam) vaccine. Others included water emulsion antibiotic drug such as CTC® 10% against Fowl typhoid. The vaccine and drugs were procured via local area extension agent. Control of ecto-parasites mainly 'Paazi' was reported to be done by majority of the respondents (n=42, 97.7%) interviewed as shown in Figure 3. Most of the farmers mentioned to use household insecticides i.e. Runguâ(Imiprothrin, 0.2 %w/v and Cypermethrin, 0.02%w/v) and Ngaoâ (16.2%) although few also mentioned others such as Carbarly group of acaricides- Akheri powderâ or Sevinâ, and water emulsion acaricides such as formamidine group i.e. Amitrazâ(23.2%). Despite of the reported acaricides used, none was reported to be effective in controlling 'Paazi'. Possible reason could be due to irregular use, unsupervised, under dosing and the scavenging nature of local chicken making a complete confinement of chicken impractical. Other control practices mentioned to be used include burning a chicken pen (32.2%), fumigating it by using hot water and ashes (4.6%). In addition to the above practices, the use of kerosene (9.6%), hand dressing infested chicken (2.3%), using palm oil (2.3%), plastering chicken pen wall in order to fill in crevices/cracks (2.3%) and the use of local herbs (ethno vet)(2.3%) was also mentioned to be used regularly in the surveyed villages. However, none of the control practices was mentioned to be effective in controlling 'Paazi'. The notable use of traditional medicine and techniques to control ecto-parasites was due to its low cost, local availability and easiness of application. This indicates that traditional medicines do work and have the potential to improve the health status of village flocks. Hence, there is a need for research to determine their chemical properties, concentrations and mode of application.
|
Figure 3. Types of control practices reported to be used in Mkinga and Maramba divisions to control 'Paazi' |
The respondents were asked to mention different ways through
which local chicken production could be improved including ranking
in order of priority. Out of 43 respondents interviewed, 35 (81.3%)
reported different ways local chicken can be improved in Maramba
and Mkinga divisions. A number of management strategies were
mentioned. These included availing drugs or medicine at farm level
(62.3%), provision of good extension services (20.9%), improving
husbandry and feeding (14.2%), provision of capital loan to expand
chicken project (1.3%) and improving genetic potential of local
chicken through the use of improved cockerel (1.3%).
This study shows that FRVC production system (which form part of smallholder agricultural system) in Mkinga and Maramba divisions is an important undertaking.
The chickens play a key socio-economic role and largely contribute to household income and protein malnutrition alleviation.
Scavenging was the major feeding system and dietary supplementation was irregular and inconsistent.
Mortality, housing, ectoparasitosis and predation were identified as the major constraint to village chicken productivity.
Health provision to chicken was limited to few farmers. Most farmers lack veterinary and extension advice.
There is an urgent need for the government to support
free-ranging poultry farmers by providing subsidized vaccinations
and technical support in order to develop and stimulate economic
development in impoverished rural areas of Tanga
region.
The authors wish to acknowledge the cooperation of farmers who
participated in this study, Muheza (now Mkinga) district field
staff for their considerable support and help. This paper is
published with the permission from the Director of veterinary
services in Tanzania. This study was funded by Eastern zone client
oriented research (EZCORE)- animal health research support grant
(Grant No 4/2006/MHZ).
Anonymous 2002a Annual Report, Ministry of Water and Livestock Development, Dar-es-Salaam, Tanzania Mainland. pp 20-22.
Anonymous 2002b Diagnostic Participatory Rural Appraisal report for smallholder poultry keepers in Mkinga and Maramba Divisions, Muheza, Tanga, Tanzania, EZCORE internal report, July 2002
Conroy C, Sparks N, Chandrasekaran D, Sharma A, Shindey D, Singh L R, Natarajan A and Anitha K 2005 The significance of predation as a constraint in scavenging poultry systems: some findings from India. Livestock Research for Rural Development. Volume 17 Article #70 Retrieved March 13, 2007, from http://www.cipav.org.co/lrrd/lrrd17/6/conr17070.htm
Smith A J 1992 Integration of poultry production into the agricultural systems. In The Tropical Agriculturist: Poultry. CTA Macmillan, pp. 179 - 191.
Swai E S, Karimuribo E D and Kyakaisho P 2007 Further evidence of occurrence of Argas persicus Oken 1881 in free-range village chickens in Tanzania. Livestock Research for Rural Development. Volume 19, Article #6. Retrieved March 13, 2007, from http://www.cipav.org.co/lrrd/lrrd19/1/swai19006.htm
URT, United Republic of Tanzania 2005 National Strategy for Growth and Reduction of Poverty (NSGRP): www.tanzania.go.tz/mkukutaaf.htm Last visited 4th March 2007
Walker A R, Bonattour A, Camicas J J, Estrada-Pena Harok I G, Latif A A, Pegram R G and Presto P M 2003 Ticks of domestic animals in Africa: a guide to identification of species. Newsletter on ticks and tick borne diseases of livestock in the tropics version 29, Feb 2006, Available on-line at http://www.icttd.nl Last visited 14th March 2007
Received 17 April 2007; Accepted 24 May 2007; Published 3 August 2007