Participatory Rural Appraisal of Dairy Farms in the North West Province of Cameroon

P H Bayemi, M J Bryant^*, D Pingpoh, H Imele, J Mbanya, V Tanya, D Cavestany^**, J Awoh^***, A Ngoucheme, D Sali, F Ekoue^****, H Njakoi^***** and E C Webb*****

Institute of Agricultural Research for Development, Bambui B.P. 51 Bamenda , Cameroon.
hbayemi@yahoo.fr
*University of Reading, Department of Agriculture P.O. Box 236, Reading England
M.J.Bryant@reading.ac.uk
**Instituto Nacional de Investigación Agropecuaria, INIA La Estanzuela, and Department of Reproduction, Veterinary College,
Lasplaces 1620, 11600 Montevideo, Uruguay
dancaves@internet.com.uy
***Dschang University, faculty of Agriculture
****Institute of Agricultural Research for Development, Yaoundé, Cameroon.
*****Heifer Project International, Bamenda Cameroon.
Department of Animal and Wildlife Sciences, University of Pretoria, South Africa, 0002,
E-mail: edward.webb@up.ac.za

Abstract

Introduction

Cameroon human population is growing at a rate of 5% per year (MINEPIA 2002). There is unceasing worry to feed this population. Annual per capita of milk production in Cameroon was estimated at 5.1 kg (MINPAT 1986) while consumption was estimated at 10kg / person / year by Von Masow (1984). Total domestic production of milk was 50,000 tonnes (Tambi 1991). In 1999, per capita production stood at 12.8 kg while per capita consumption was 15.3 kg in 1998 (calculated from FAO 2000). Milk production in the country has substantially increased in the last 15 years from 48,000 tonnes to 184,000 tonnes. However, the production is far from satisfying local demand for milk and milk products. This gap in domestic demand is being over the years covered by large imports. Teuscher et al (1992) estimated that the level of imports of milk and milk products was 11480 tonnes, which represented about 50% of the adult per capita consumption. However, due to the devaluation of the CFA Franc currency used in the country by 100% in 1994, per capita consumption in subsequent years dropped to less than half of Africa's which is 34 kg/person/year (294kg/person/year in Europe) reflecting the limits in imports of dairy products in the country, standing only at 23% of total per capita consumption. Imported products are expensive for common Cameroonians. Consequently, products from local milk can more efficiently compete with imported ones. This paves the way for a huge development in local milk production. This development will be effective if there is a detailed knowledge of the dairy production environment in the country. Interventions will be easily measured, felt and seen if they were applied in selected farms and economic effect of interventions monitored.

More than ten years ago (Douffissa 1988), three cattle systems were defined in Cameroon: traditional, semi intensive and intensive. However, the economical, political and environmental conditions have drastically changed in the country. There is limited knowledge of current milk production systems in Cameroon. What are their constraints and limitations? What stakeholders and factors influence dairy production? What is the marketing channel? What perceptions have dairy farmers in this activity and what suggestions can be given to improve dairy production in the country? What is the economic reference point of selected farms that could be used for measuring results of subsequent interventions? The following research was conceived to attempt answers to the above questions.

Material and methods

Participatory rural appraisal (PRA)

The Participatory Rural Appraisal was used as a tool to identify constraints and suggest interventions geared towards promoting the sustainable development of dairy production in the North West Province (NWP) of the Western Highlands Cameroon.

Choice of area

This site was chosen because it is the most appropriate environment for dairy improvement in the country. Not only it is free of Tse Tse fly, but the region is linked to the two major towns, Douala and Yaoundé by a good road network. Temperatures are the lowest in the country therefore suitable for high yielding breeds. Milk production in this area is government priority. A dairy technology laboratory has been set in the area to improve milk processing. Furthermore, the Heifer Project International has established its head quarters in the region to closely help farmers in dairy management. On the other hand, milk produced in the northern part of the country is mostly through pastoralists who are not very open to change in cattle management. The North West province is therefore very suitable for improvement in dairy production and was chosen consequently.

Research team and PRA method

A multidisciplinary research team was constituted with scientists of different fields as follows: a cattle and forage scientist, an agro economist, a dairy technologist, a veterinarian, an extension agent and various technicians. One researcher, a lady helped to establish links with women in communities were foreign men were not allowed to individually question women. The team first decided on the site and farmers' group to be visited. Contact was established with the group through phone numbers or government 'zonal extension workers'. A date and time were arranged at the convenience of the farmers' group. Some food or local drinks were prepared to make the discussion free and informal. A PRA topical guideline was prepared as a semi-structured interview consisting of the following points: introduction on purpose of the meeting and visit; presentation of the team; presentation of individual farmers present; questions on cropping, cattle management, milk processing and marketing, labour, constraints and prioritization, other activities, expectations, other comments. Finally a dictation machine (tape recorder) was discretely used during group discussions to avoid writing down every piece of information. The questionnaire was translated in the very common pidgin language by a researcher, native of the area. It was pre-tested at one locality (Sabga) and readjusted for shortening the time taken for group discussions, order of questioning and few more questions were added. Sometimes there was a need to translate part of the discussion into/from local dialects. This was either done by a research team member or a member of the dairy group.

At arrival at the site the research team was welcomed by the extension agent who introduced it to the group. The team explained the purpose of the visit and initiated the discussion. The team sat in an indiscriminate manner, and mingled with farmers to create more confidence among both parties. Although a leader guided the discussion, any other scientist could intervene at any moment to ask for appropriate information that could have been overlooked. The zonal extension agent, key informant, also took part in the discussion. He gave additional information that the farmers were unable to provide. Care was taken to avoid people monopolizing the discussion because of their wealth status or leadership role.

The group discussion proper took place in one of the farmers' houses or in a common group house. It lasted a maximum of two hours. After the discussion a field visit to various farms was organized. In some cases the research team divided itself into sub groups of 2-3 people; in other instances the whole team visited farms. Much information was also gathered from questions asked to farmers during the village walk. Transects and direct observations were done during the field walk. Body maps were drawn by 36 farmers to diagnose health problems in 2 groups .Visits usually ended with a meal taken with farmers. However, there were further questions asked to the key informant to collect additional information or crosscheck farmers' answers. The visit lasted a day in each group. In some villages (Sabga, Jakiri, and Tadu) because of gender-separated way of life, women were interviewed from men in separate visits. A total of 137 farmers were individually interviewed representing about 25 % of dairy farmers registered in 32 dairy groups and 3 dairy cooperatives. Interviews were also carried out with other stakeholders of the dairy sector such as Non Governmental Organizations, industries and government officers. The research team met the day following the visit to gather information and write a report.

Secondary data used in this work involved annual reports of IRZ (1984, 1985, 1986); results of previous surveys involving dairy production in the region (HPI 1999; Kameni et al 1999); current statistics from dairy cooperatives and non governmental organisations.

Data analysis of PRA information

As suggested by Pretty (1994), the validity of the data was ensured by: forming a research team with members having differences in scientific disciplines, ethnicity, age, religion and gender. Main points noted in a group were read to the participants at the end of gathering and corrected. During the research report writing, there was crosschecking of information collected by each member of the multidisciplinary team. The results were submitted to review in a scientific presentation to colleagues and to peer review. Results were subsequently compared to secondary data.

Three methods were used for constraints ranking: first, farmers agreed among themselves and a spokesman expressed the major constraint by order of importance. Where this agreement failed, farmers were asked to close their eyes and put their hands up when a constraint was listed. Eyes were closed to reduce influence from other farmers. The numbers of farmers voting for a particular constraint was subsequently counted. Lastly, Ashby's method (1986) was used to rank constraints in the whole region. Simple statistics were used to calculate percentages.

Results

Seasonal calendar of activities of dairy farmers

Farming activities of dairy farmers of the Western Highlands can be divided as indicated in Figure 1.

Activities such as milking, forage cut and carry for purebreds Holstein or Jersey, milk processing and marketing are permanent throughout the year, while farming and transhumance depend on the 2 seasons, wet and dry. These activities are similar for the whole region.

Cattle feeding

Much milk produced in the Western Highlands of Cameroon is from native cattle. These cattle are usually grazing native pastures of Sporobolus africanus, Pennisetum purporeum and Melinis multiflora. However, pastures have also been improved with planted grasses, legumes and multipurpose trees (MPT). These are for grasses: Brachiaria spp, Trypsacum laxum, Kikuyu (Pennisetum clandestinum). Trees are being planted as live fences and are: Caliandra spp, Leucaena leucocephala, Jacaranda spp and Acacia spp. Cattle are also fed on by-products, waste food and fruits such as sugar cane leaves, potato leaves, guavas, ripe bananas, pumpkins and waste cooked corn (fufu). Crop residues include: corn stovers, banana pseudo stems and leaves, ground nuts and beans haulms.

The following ingredients are commonly used in supplementation: Maize, rice bran, wheat bran, palm kernel cake, cotton seed cake, whole soya beans, bone ash, limestone and table salt. The proportion of farmers using different feedstuffs in the peri-urban areas of Bamenda and Fundong is shown in Table 1.

Principal energy providing ingredients are maize and wheat bran while protein mostly comes from cotton seed cake or whole soya bean flour. Traditional farmers use table salt as sole supplement to cattle in the areas of Sabga, Jakiri and Bamdzeng. Concentrate is fed to dairy cows during milking.

Breeding

Traditional producers have been using for a long time natural mating from visually-selected bulls bought, exchanged or loaned from other farmers. With the aim of upgrading the traditional stock, artificial insemination (AI) is being used in Jakiri, Tadu and Bamdzeng. AI is also used in Holstein cows in Fundong (Meli and Mukweh), Nkwen and Santa. In the Bamenda surroundings where there is a high failure of AI, purebred Holstein bulls are used for natural mating. In Sabga and Jakiri, farmers use crossbred bulls for breeding therefore obtaining ¼ Holstein crosses in the progeny.

Dairy cows are bred as from two months post partum. The calving interval lies between 12 and 18 months. Improved breeds reach active sexual maturity at 24 months while local breeds are sexually active as from 36 months of age.

Calf management

Calves from purebred dams are weaned at 4 months of age while in traditional herds weaning is between 7 and 12 months. Weaning is done by separating calves from cows for 3 weeks or by rubbing a mixture of rotten colostrum and dung in mother's udder, thus repulsing the calf. Calves from local breed suckle dams for one month before milking starts. Purebred calves are first given colostrum; then are bucket fed in elevated pens as from three days post partum as follows: less then one month 5-6 litres of milk/calf/day; 1 to 2 months, 3-4 litres/calf/day; 2-3months, 2-3 litres/calf/day and 3-4 months, 1litre of milk/calf/day. Forage is introduced between 2 and 4 weeks of calving.

Milking

In traditional herds, cows to be milked are chosen with the following criteria: calmness and high milk production, older cows are preferred to young ones. In general milking is done in the morning between 6 and 8 am and in the evening between 4 and 6 pm. Purebred Holsteins are milked twice a day. In Tadu milking sometimes goes up to 3 times a day. Milking of local breeds is still done with poor hygiene consisting of few seconds suckling by calf to favour milk let down (in fact, the contrary is the case as restricted sucking results in lower somatic cell counts and less mastitis: the Editor). Milking in then done with calf presence. Dairy farmers owning purebred Holstein clean the udder with warm water and rub it with Vaseline before and after milking. Milking is done manually and milk is collected in buckets after tying the cow's hind legs.

Production

Breed production is summarized in Table 2.

The impact of pure Holstein Friesian is increasing in the region, particularly in peri-urban areas of Bamenda in Mezam and Fundong in Boyo (Figure 2). Many women are also involved in rearing this breed.

Figure 3 indicates the quantity of milk collected by Sotramilk (see below). Much milk is produced in the rainy season when forage is abundant.

Figure 4 shows that the sources of milk products in the country are either imported or locally produced.

Milk not taken by calves is either home consumed, spoilt or marketed. No cooperative collects milk. Milk and milk products are marketed either directly by producers or are collected by processing companies. In the peri-urban area of Bamenda, milk marketing is more formal through Sotramilk. In other areas, milk marketing is informal. Milk is either bought at the farm by individuals or farmers carry it to the market place. In this case the price is higher than what is offered by the processing plants. This pattern also happens in the northern part of the country with much activity where former processing plants closed up because of lack of sufficient milk supply. Costs of products are indicated in Table 3.

Home processed products and milk sold in the open market are still much a part of the marketing channel. If better quality products are to be supplied to consumers, there is a need for methods on processing and hygiene to be extended to farmers.

On average, 20 to 50% of milk is home consumed in places where marketing is a problem. In Meli (Fundong), sour milk (spoilt) is sold to dog owners. After milking, women and children can take up to 2 hours to get to the market place, and wait for up to 4 hours for the milk to be sold before going back to the homestead.

Housing and manure

In traditional management, cattle graze on a free-range system with nights spent in open pastures. In the peri-urban management, there are sleeping paddocks and sleeping sheds constructed with wood and the roof made of either zinc material or local grass. In the last system cows are milked in sheds cemented with concrete. In the zero grazing system, some farmers had very poor sheds letting mature cows eat grass from the roof, thus destroying it.

Manure is a very important activity in the cut and carry (zero grazing) farms and in peri-urban areas where gardening and agriculture are much rewarded. This manure is collected with spades and kept in pits before being dried, used or sold. There is little hazard from manure as the quantity is small.

Labour and gender

Generally, the whole family is involved in cattle caring. In the more pastoralist communities, adult men take cattle for grazing while women and children do the milking, processing and milk marketing. School children milk cows before going to school in the morning. The peak period for labour demand is the dry season (January to March) when the farming season starts. At this time, labour is hired to take cattle for transhumance at a cost of 15 000 FCFA per month. Labour is paid for farm preparation at 10000 to 20000 FCFA for a one eighth hectare farm or 600FCFA per man day. When men are married to many wives, at least one of them is in charge of farming and gardening. In the Fulani pastoralist communities, men own cattle but the milk belongs to the women. In Mukweh, some labour is paid with liquid milk. In Sabga peak labour demand in the rainy season is for training of first-calf cows for milking.

Individual interviews revealed that men constituted only 16.8% of dairy operators compared to 83.2% of women. 71% of respondents fell within the age range of 24 to 50 years while 29% was above 50 years. Furthermore 49% of farmers had not received training in dairy production.

The gender representation of dairy farmers in the North West Province is illustrated in Figure 5.

The high percentage of the Fulani women clearly shows that they are more involved in dairy activities than the None Fulani women. Thus, the role of the women in dairy production is of prime importance. They need motivation and more education in this sector. If the milk production sector fails, then the livelihood of the Fulani women will be the most affected.

Health

Major dairy cattle diseases in the Western highland are in order of importance ticks and tick born diseases: babesiosis, anaplasmosis, dermatophilosis, cowdriosis; mastitis in milking cows; diarrhea; foot and mouth disease (FMD); black quarter; ephemeral fever and ear infection. Veterinary services are provided by non governmental organizations or private veterinarians. Vaccination is done yearly against black quarter, haemorrhagic septicaemia, and contagious bovine pleuropneumonia. Only few farmers spray their dairy animals. Hand de-ticking is more common. It is the fear of ticks that prevents some farmers from sending crossbred cattle in low and hot lands on transhumance during the dry season. Many traditional farmers make use of ethno- veterinary medicine (Sabaga, Jakiri, and Bamdzeng).

Marketing

Stakeholders

Main stakeholders for dairy production in the Western Highland of Cameroon are: The non government organization Heifer Project International (HPI), the processing company Sotramilk, the Tadu dairy cooperative, the Ministry of Livestock and Fisheries, feed retailers and the Institute of Agricultural Research for Development.

HPI

This non governmental organization (NGO) is based in the United States of America. In Cameroon it is a representation of the intensive form of management. In 1974, in collaboration with the institute of animal research (IRZ), HPI provided the initial shipment of purebred Holstein and Jersey dairy cattle to Bambui research station. Recently, more purebred in-calf heifers are being provided directly to farmers.

Sotramilk

It is a plant aided by a Dutch non governmental organization which started operating in 1995 with the aim of boosting local milk production. It operates by buying milk from neighbouring farmers in a radius of 10 km. The factory has a capacity for processing 12000 litres of milk a day. But it collects only a maximum of 300 litres per day in the dry season and 600 litres per day in the rainy season. Therefore this fresh milk is usually combined with imported powder milk to make various products such as: Gudali and Edam cheese, and natural and fruit yoghurt (cherry, pineapple). Milk collection is limited because of bad roads. However farmers who do arrange for their milk to be delivered to the factory are compensated for transport.

Tadu Dairy Cooperative Society (TDCS )

TDCS is a cooperative organization established in 1992 following an intensive training provided by Land 'O' Lakes, Inc. and the United States Agency for International Development (USAID). The main service provided to the members is artificial insemination with Holstein and Brahman semen. TDCS was initially made up of over 200 pastoralists. Up to 1000 to 3000 liters of milk could be collected daily during the dry and rainy season respectively. However the road network is very bad and the market is limited to the neigbouring Kumbo town. Consequently TDCS is planning to build a processing plant.

Feed retailers

There are many feed retailing shops in provincial towns. The following are ingredients are commonly found in these shops: cotton seed cake, wheat bran, rice bran, soya bean cake, fish meal, palm kernel cake, bone ash, limestone meal, blood meal (Table 4). Maize is usually sold for human consumption although it is also bought by animal producers.

Use of money from sales of milk

In Fulani communities (pastoralist tendency), milk belongs to the women and so money from milk sales is not used for cattle but for personal needs, household needs and children school fees. In native communities, this money can also be used for cattle. Some people suggested that if they had more money from milk, they would send their children to secondary school and university or open other businesses.

Farmers' access to services and information

Farmers acquire information and services mainly through NGOs and the government programme of agricultural extension services. Although internet services are available in the regional head quarters, not many farmers are formally educated to be interested to this source of information. What is needed is to provide adequate information to these organizations for it to be well used by farmers. Therefore if the relationship of research continues to be good with these stakeholders, there will be a good flow of information. Most of the veterinary services are provided by HPI as far as Holstein cows are concerned. Otherwise, private veterinarians offer payable consultations.

Other activities

Dairy farmers commonly keep other farm animals such as scavenging chickens, goats, sheep and horses. Non-Muslims keep pigs. All farmers visited plant crops, and grow maize, potatoes, sweat potatoes, coco yams, beans or coffee. Some farmers are also involved in activities such as cattle trading, small businesses and apiculture. Some Fundong dairy farmers own fishponds.

Constraints

Major constraints of dairy production in the region were ranked by order of importance as follows:

Marketing: Poor marketing opportunities and long distances to market.

The processing company Sotramilk has a limited sphere of milk collection because of high collection costs. Also, farmers supplying milk to the factory complain of the price of fresh milk being low.

Farmers' suggestions to alleviate constraint: Break the monopoly of Sotramilk or let the company raise prices; let the government provide financial support to farmers.

Recommendation from research team: to this constraint, farmers were advised to arrange milk collection in groups and transportation to Sotramilk.

Farmers of Fundong and Bafut who have adopted this system no longer suffer from a marketing problem.

Milk transport is one of the most important problems faced by farmers. In the traditional system with large herds, individual cows have a low daily production but farmers can milk a large number of animals to have a good quantity of milk per herd. Unfortunately, these farms are far from the market and roads are bad thus milk easily spoils during transportation. The Lacto-peroxydase system could be used to prolong the shelf life of milk. Farmers could also be provided with technologies for processing milk into cheese if the market is available.

When considering the peri urban production system, the most important constraint was the lack of improved breeds for milk production (Holstein Friesian).

Feeding: limited grazing land and poor supplementation of cattle

Because of encroachment of pastures by crop farmers, dairy farmers found themselves with limited grazing lands. They also lack knowledge of good feed compounding. Some areas (Mukweh in Fundong) had their improved pastures permanently invaded by stray goats from neighbours. In Tadu most pastures are seriously invaded by bracken fern. There was consequently an inadequate feeding during the dry season in most parts of the region. Moreover, farmers find the cost of concentrate high and the cut and carry system tedious.

Farmers suggested as a solution to land conflicts that the government secure communal grazing lands.

Recommendation from research team: Make concentrates more available at low cost by formulating in a linear programme cheap rations using local available products. Research should also find a way of making grass cutting less tedious for farmers employing zero grazing (cut-and-carry) management. This could be done by devising a local chopper adapted to the farm size and which could be acquired by dairy groups.

Previously, dairy rations formulated by researchers in the regions, though efficient, were not very profitable because of inclusion of high cost ingredients. This constraint can be alleviated by extending improved and cheaper rations to farmers, and providing them with extension leaflets in feeding methods for dairy cows.

Limited health control

In some places, drugs and veterinary services were not always available.

Recommendation from the research team:  Farmers should invest more money in health control and should organize themselves within dairy groups for veterinary care in order to lower cost. Private veterinarians are available for consultation.

The government no longer provides free veterinary services to farmers as the sector has been liberalized. It is good to make farmers understand the economic loss they incur with poor health control. The Economic Opportunity Survey will be helpful in this area.  Previously, veterinary researchers worked to identify diseases and pathogens hindering cattle productivity. No economic evaluation of the impact of disease has been done in the region.

Inadequate knowledge in processing, conservation and storage of milk

Many farmers complained of their milk getting bad in a short time.

Recommendation from the research team: Train farmers in hygiene, processing and preservation methods. The lacto-peroxydase system, tested in the Sabga, Santa and Bali region, can extend the shelf life of milk to an additional 6 hours or more if the milk is kept in cold water. This constraint will be alleviated by extending processing and conservation methods.

Breeding: poor reproductive management and prolonged calving interval

In traditional management where fencing is rare, bulls breed cows without control. This was the case in Jakiri and Mbamzeng. On the other hand, farmers using AI did not always have semen or purebred bulls available to them. There was also a great failure of conception in inseminated cows.

Recommendations: Improve the breeding program. This constraint though 5^th in general ranking, comes among the first in the peri urban system which is the most market-oriented. Farmers complained of lacking good dairy breeds. The causes of the poor reproductive performance need to be investigated and advice given to farmers. There is also a need to emphasize good breeding management through extension leaflets.

Artificial insemination has proven to be less sustainable because of the high cost of liquid nitrogen and conception failure due to the AI technician not getting 'heat' information early enough or getting to the farm at an appropriate time. Therefore, it is advisable that dairy groups acquire bulls. It is also possible in areas where the main problem is high cost of liquid nitrogen that fresh semen be used for insemination.

Farming management

Lack of water in the dry season
Recommendation: Farmers could pay people to carry water when they use a zero grazing scheme.

Poor Housing

Many milking stables were in a bad state and fencing was poor
Recommendation: Provide extension services for housing and fencing.

Poor organization of group

Recommendation: Reorganize groups to be more dynamic.

Others: limited number of dairy cows. 

Poor record keeping

Dairy production systems

Dairy cattle management systems of the region are summarized in Table 5. 

In the region, five main dairy production systems have been identified.  Recommendations from stakeholders are indicated in Table 6. 

Discussion

Feeding, breeds, production, marketing and health

The calendar of activities of the region shows that after the harvesting season, crop residues can be fed to stall-fed cattle in the dry season or as the practice is in mixed communities (pastoratists/crop farmers), cattle can be sent to graze standing crop residues such as maize stovers. In Kenya, Stall at al (2001) found that crop residues (maize stovers) used for dairy cattle had gone up while the use of concentrate and roadside grasses was reduced. Chopping grass for Holstein cows is tedious in the zero grazing system. These farms could have a tremendous help if research could design a small scale chopper to be used by farmers' groups.

All dairy systems in the North West province are involved in cropping activities; although in varying degrees as described in Figure 6. 

It is expected that farmers more involved in cropping will feed more crop residues to cattle. Therefore, they may not have severe feeding shortages in the dry season. On the other hand the fear is that these farmers rely more on the residues at the expense of grass. It is therefore necessary to keep encouraging crop farmers to cultivate improved grass and legumes.

The information on ingredients used in concentrate shows that most farmers use maize and whole soya bean flour in the feed. But this seems expensive as there are by-products such as wheat bran and cotton seed cake which can more economically be used. Palm kernel cake is cheap but is not widely used as farmers complain of a drop in milk yield when including it in high proportions in home-made concentrate. This by-product is not very palatable (McDonald et al 1988) and has residual oil content which can impair microbial activity of the rumen. Therefore, the recommendation of using a linear program to formulate dairy rations seems appropriate.

The only high yielding pure milk breed found in the region was Holstein. In the past, the smaller size Jersey was also present. Farmers of the Western Highlands of Cameroon prefer Holstein Friesian which is also good for beef production. The increasing use of the Holstein Friesian breed indicates that in market oriented farms, farmers prefer high yielding animals to improve milk production per cow. This confirms their willingness to adopt new technologies. A company or organization set up to produce these animals will be of great help to the farmers. In Kenya (Stall et al 2001), although the predominant dairy breed is Holstein (42%), there are also other exotic milk breeds: Ayrshire (18%); Guernsey (12%), Jersey (3%); the local breeds representing 25% of milking animals. In Cameroon, production from crosses is fairly good. These animals are well adapted to the semi intensive and to the improved extensive management. Twice daily milking was found to increase milk production by over 30% compared to once-daily milking but the market in many places is not good. That is why many farmers do not do milk more than once daily.

Main factors influencing milk production have been outlined in Figure 7. Fresh milk price can be increased by reduced importation. This is in fact what has happened since 1994 because of the devaluation of the CFA currency. The milk price can also be put up by processing plants. 

Lastly in open markets, a high consumer demand increases price. When the milk price is up, farmers can modify their management by improving it. They can also reduce culling rates of female dairy animals. Others can purchase high yielding animals which will help them make more money. In contrast, when price is low, culling rates go high in favour of meat and management is poor. Processing plants sometimes modify eating habits of Cameroonians by creating new products. When consumer demand for fresh milk and milk products is high, milk production can increase because many people purchase products directly from farmers. As in the first case, the need of increased production at farm level positively influences type of breed kept and management. Research and extension bodies intervene mainly in breeds, breeding and management. At the moment, there is no possibility of over production which could negatively influence price. This confirms the findings of Tambi and Vabi (1994) that price is relatively inflexible to changes in market supply. The demand for milk is very high in urban areas and there is good prospect for farmers if production increases. Research and extension consequently have a lot to do in improving management, especially in the peri-urban system.

The pattern of milk supplies to the processing plant indicate that there is less milk available in the dry season. Consequently the plant must import powder milk at that period to meet a high demand in milk products. However, the plant could encourage farmers to keep up with the production in the dry season by paying a higher price of milk. The informal marketing could be improved by farmers' groups organizing themselves to preserve milk and market it together.

As far as consumption is concerned, many people do not know the importance of milk in human nutrition. Fresh milk is not also common in eating habits of non pastoralist Cameroonians. If many farmers do not sell their milk, it is simply because of the lack of market. In Kenya (ibid) 37% of dairy farmers' households refuse to sell their milk. In Cameroon, most dairy operators are women. This is an indication that women constitute a stronger force in milk production and marketing than the men do. This agrees with the findings of Kameni (1994). This also occurs in Zanzibar, Tanzania where 75% of them are women (Biwi 1992). This fact is a good prospect for women's welfare in rural Cameroon. If dairy production increases, women will be the first beneficiaries. They need motivation and more education in this sector. Otherwise, their livelihood will be most affected.

In health aspects, worm infestation was not listed as a problem. Perhaps because much ethonovetenary drugs are used to control worms (Nfi et al 1999). It is also possible that worm infestation causes much morbidity thus reducing milk yield. It would be better for farmers to make sure that worms are controlled even if cows do not clinically suffer. Most important diseases were found to be tick-born; meanwhile in Nigerian dairy commercial farms, it is mastitis (Onwuka et al 1995). In the semi-intensive system, specialized dairy breeds are sent for grazing at the time tick infestation is high in pasture (Bayemi et al 1999). Therefore there is a need to control this challenge with pour-on acaricides so that exotic breeds or crossbreeds are not put at risk.

Systems

Nell (1992) described 5 dairy production systems in Sub Saharian Africa:  pastoral, agro pastoral, mixed farming, intensive and peri urban. These systems are found in Cameroon in various stages. However, the intensive system if understood as leading to industrial milk production was not found in the western Highlands where most farmers practice dairying at a small scale level. Previously only three dairy systems were found in Cameroon: pastoral, semi intensive and intensive (Kameni et al 1999). Over the years, with the growing extension of high yielding dairy breeds, the zero grazing system has been encouraged. In the past, this intensive system was not encouraged as there was a fear that farmers would not be able to care for heavier animals such as Holstein. In fact, this is what happens in some of the zero grazing farms where farmers,  because of the lack of market,  neglect these animals. However, in the peri-urban area, many farmers adopt good management and would like to increase the number of Holstein cows in their farms. Dairying has also specialized in the peri-urban areas where marketing is good. More crop farmers get involved in dairying. In Tunisia, the total confinement system is practiced by farmers without pastures (Lahmar et al 2003). But in Cameroon, many such farmers cultivate improved grass in a cut-and-carry system.

In Tanzania (MOAC 1998), four systems are found in small scale dairying: two intensive urban systems and two semi intensive urban, with manure being very valuable alongside milk or meat. Farmers in that country also use crosses in the intensive system while in the North West of Cameroon, crosses are only used in the semi-intensive. The semi zero grazing found in Uganda (ILRI 1996) or Kenya was not found in this study.

Constraints

In Uganda, the primary constraint was seasonal fluctuation in quantity and quality of feed resources. This is similar to Kenya and Tanzania. This must be related to the fact that dairying is more developed in those countries and dairy cooperatives well established. In Cameroon, the first constraint was market related when considering the whole region. A lot more needs to be done to channel the milk produced in rural areas to urban centers and educate more people in the use of milk in their diets. This can be done by processing plants, cooperatives or NGOs. However in the semi intensive and zero grazing peri urban systems where the problem of marketing is reduced, farmers complain of the lack of good dairy breeds. This shows their desire to improve production. Their willingness to invest in the purchase of high yielding dairy cows is an indication that small scale dairy business is profitable.  In Zimbabwe (Francis and Sibanda 2001), the first constraint was poor reproductive performance followed by inadequate amount and low quality forage. This pattern is similar to the peri-urban systems of the northwest province.

Main constraints of dairy production in this region could be tackled through extension of appropriate methods adapted to each system. In fact, recommendations from farmers show that government, research and extension services are keys to the solution to many of these problems

Conclusion

• Five dairy production systems were found in the North West province of Cameroon: transhumance, improved extensive, semi-intensive, peri-urban, and zero-grazing.

• Constraints hindering production in these systems are by order of importance related to marketing, reproduction, processing and milk preservation.

• In market-oriented farms, main constraints are poor reproduction and inadequate feeding.

• Main factors influencing production are milk processing plant, consumer demand, fresh milk price and genotype/management. The marketing channel involves processing plants and open markets.

• Recommendations were geared towards government bodies and research working more to helping farmers to resolving constraints. However, processing plants were also advised to finding ways of establishing more confidence with farmers in the measurement of their milk.

• The fact that women are the gender more involved in milk production and benefiting more of the marketing is a good prospect for their welfare if production improves.

• The most urgent interventions to follow in market oriented farms seems to be tackling constraints of failure in reproduction and heavy work load in chopping grass. This is coupled with the extension of improved methods feeding and breeding methods.

Acknowledgments

This work was carried out in the framework of a Coordinated Research Project sponsored by the International Atomic Energy Agency (IAEA). The authors are grateful to Dr Devendra for correcting the manuscript.

References

Ashby J A 1986 Methodology for the participation of small farmers in the design of on-farm trials. Agricultural Administration 22 (1-9)

Bayemi P H, Ndi C, Nfi A and Ekue F N 1999 Localization and seasonal dynamics of ticks (Ixodidae) on trade cattle of the North West region of Cameroon. Bulletin of Animal Health Production in Africa, 133-137

Biwi K M 1992 Development of small holder dairying in Zanzibar Future of livestock industries in East and Southern Africa. Proceedings of the Workshop held at Kadoma Ranch Hotel, Zimbabwe, 20-23 July 1992, pp 173-176

Douffissa A 1988 Production et commercialisation du lait dans le Mbéré: Données complémentaires. Contact Volume 5

HPI (Heifer Project International) 1999 Evaluation of the dairy program of Heifer Project International (HPI) in Cameroon; Report, Bamenda Cameroon

International Livestock Research Institute (ILRI) 1996 The Ugandan Dairy Sub-Sector. A rapid Appraisal pp i.

IRZ (Institute of Animal Research) 1984, 1985, 1986 Annual Reports Yaoundé, Cameroon

Kameni A 1994 Some aspects of the production of Bafut cheese PhD Thesis, University of Reading, United Kingdom

Lahmar M, Bouraoui R, Lachaal L and Toumi J 2003 Characterization of milk production systems in Tunisia. The case of Ariana Region. Editeurs: Djemali M, Guellouz M, Ariana, Tunisia, pp 188-192

McDonald P, Edwards R A and Greenhalgh J F D 1988 Animal nutrition 4^th edition Longman Scientific and Technical, Essex, UK 462 pp

MINEPIA (Ministère de l'élevage, des pêches et des industries animales) 2002 La stratégie sectorielle de l'élevage, des pêches et des industries animales. Cabinet management 2002, Yaoundé