3. FINDINGS

This chapter presents the summary of the findings of the PRA sessions, interviews with key informants and case studies carried out with selected fish farmers. More detailed results are contained in Appendices 1 to 5.

3.1 General background of fish farming in Subira and Ihanda Villages

Fish farming was started in Subira in 1961 by a villager who captured the first fingerlings (species unknown) in Ruvuma River which runs through the area. Over the years, other farmers copied the activity. In 1976 a fisheries expert visited the village and encouraged the farmers to construct more ponds. In 1990 the village was visited by a US Peace Corps Volunteer who gave advice on various fish farming activities. The District Fisheries Office provides extension service for fish farmers visiting the office in Songea. In 1993, there were 45 active fish farmers out of a total of 375 households. The village covers approximately 16 km².

Table 1: Climatic, agricultural and demographic information for Ruvuma and Mbeya Regions.

Source: Atlasi Mpya Kwa Shule Za Tanzania. Bureau of Statistics: 1988 Population census.

Subira receives an annual rainfall of 1125 mm (Table 1). The rainy season covers the months from November to May with March having a profound peak (see Appendix 1.A). Most ponds are situated close to the rivers and streams which have water throughout the year. Only a few ponds are fed by water from springs. The soil in the area is generally sandy with a top layer of clay on the river banks. The species grown is mainly Oreochromis niloticus, although a number of wild tilapia as well as catfish (Claris gariepinus) also enters the ponds.

In Ihanda a few farmers constructed the first ponds in 1969 after having been inspired by a farmer in a nearby village. But it was not until a Peace Corps Volunteer (PCV) was based in the village in 1985–87 that a larger number of farmers took up the activity. Another PCV followed in 1987–88 but his contribution is not perceived to be as substantial as that of his predecessor by the fish farmers. The District Fisheries Office continued the extension service by having an Assistant Fisheries Officer based in the village in 1989–91. Currently, fish farming is carried out by 22 fish farmers (3%) of the village's 740 households. Respondents felt that they are not able to satisfy the demand for fish in the village. A considerable number of dried fish from Lake Rukwa is sold in the village, which covers roughly 25 km².

The southern part of Mbeya Region where Ihanda is situated has a climate marginally suitable for tilapia fish farming. Temperatures vary between 16 and 21°C and annual rainfall is 880 mm. There is one rainy season stretching from November to April (see Appendix 1.B). Water runs in streams and rivers throughout the year. The ponds are confined to these areas. The soil is clay mixed with sand. Oreochromis niloticus is farmed, often mixed with other species of tilapia and catfish which have come from the rivers.

3.2 Access to land

According to respondents, it is possible to acquire new land suitable for fish farming in Subira. It is usually borrowed from other villagers for as long a time as the farmers utilize it. There is no payment involved. However, the owners' right to claim the land back can obstruct long-term investments at the site. Often, new land can be obtained only at a distance from the farmers' home.

In Ihanda, agricultural land can be obtained through the village government without any cost. Farmers can also obtain land through purchase from other farmers: T.shs. 74000 (US$ 167) for a hectare of normal farm land and T.shs. 99000 (US$ 224) for a hectare of land in the river valleys are the going rates⁸. However, these river valleys to which fish farming is confined are all occupied, most often by vegetable gardening. Due to this location of the ponds, many farmers have to walk up to 45 minutes in order to visit their pond.

Fish farming in Subira and Ihanda is a predominantly male activity. The kinship system in both villages is patrilinear in which ownership, inheritance and settlement follow the male line. In Subira, there are two female fish farmers. One operates two ponds together with her husband. The other woman is a divorcee who decided to start fish farming to provide relish and even expects to have another pond constructed. A third woman -- unmarried -- has abandoned her pond temporarily because it has been flooded⁹. In Ihanda there are no female fish farmers.

While in the case of male-owned ponds the husband is the sole decision-maker on fish farming investments and activities, women and children play an important part in the daily management of the pond, i.e. feeding and manuring, as well as in the marketing of the fish. Harvesting is mainly carried out by boys and young men.

3.3 Characteristics of the fish farmers participating in the case studies

The fish farmers interviewed for the study (“respondents”) do not belong to a specific age group. The age range is 30–65 and 31–60 for Subira and Ihanda respectively, the average being 46 and 44 years¹⁰. (Tables 2A and 2B).

All the respondents have some basic education, i.e. they posses reading and writing skills. This is obtained through either primary school education or adult education. They all speak Kiswahili, but only a few, particularly those who have worked outside the village, understand English.

⁸ The exchange rate at the time of the study (August–September 1993) was US$1= T.shs. 442.
⁹ In Ruvuma Region only 2% of the fish farmers are women (Seki and Maly, 1993).
¹⁰ A more detailed description of each farmer is given in Appendix 7.

Map 2: Map of Subira village

Legend:
	Gravel Road/Foot Path		Farm Area
	River/Stream		Fish Pond Area
	Village Boundary		Granary
	Water Pipe		School
	Water Channel		CCM Party Office
	Bridge		Muslim Prayer House
	House		Church

MAP DRAWN AT PRA IN SUBIRA, 16/8–93 BY

Mr. Mohamedi Maganga
Mr. Donatus Haule
Mr. Juma Swarehe Ganda
Mr. D J Nchimbi
Mr. Peter Mhagama
Mr. Atha Ndalama
Mr. B Ndimbo
Mr. Paulo Kayumba Ndunguru
Mr. Lucus Nungu

Map 3: Map of Ilanda village

	Tarred Road		House		Grinding Machine
	Gravel Road/Foot Path		Farm Area		Shop
	Railway		Fish Pond Area		School
	River/Stream		Vegetable Garden Area		Church

	Pipeline (Underground)	MAP DRAWN AT PRA IN IHANDA, 9/9–93 BY
	Village Boundary	Mr. Remsoni R. Msongole Mr. Philemon K. Sanga Mr. Beckisoni Makama Mr. Lumuli Kayange Mr. Alison Simuchimba Mr. Eliasi Sikana
	Hill
	Dispensary

Many of the respondents hold political or influential positions as either “Ten Cell” Leaders¹¹, members of Village Government or Committees. Holding a political position may provide some contact with other institutions or extension workers.

Table 2.A. Characteristics of case study respondents in Subira Village.

¹⁾ The husband and the wife each own their ponds, but operate them as one unit.
²⁾ Farmer no. 4 is the main pond constructor in the village.
³⁾ Farmer no. 6 occasionally constructs ponds.

¹¹ A Ten Cell is the lowest level in the administrative hierarchy, comprising approximately ten households.

Table 2.B. Characteristics of case study respondents in Ihanda Village

¹⁾ Farmer no. 1 is a farm cooperative in which all tasks in fish farming are carried out after discussion and common agreement among the heads of the three households in the cooperative. In case of disagreement, however, the head of the cooperative has the final word. A boy is hired to take care of daily management of the ponds.

The interviews with the key informants revealed that in Subira as well as in Ihanda marriage between members from different clans and religions is possible. The study further disclosed that respondents have often been living and working in other parts of the country before moving to the village. These findings indicate a relatively high social and geographical mobility in the areas studied.

3.4 Income-generating activities

Farmers from both Subira and Ihanda depend on a range of activities to earn their income. Cultivation of agricultural crops is by far the most important source. In Subira, income-generating crops include maize, tobacco, vegetables (spinach and tomatoes), bananas, sugarcane, rice and sweet potatoes. Minor activities comprise sale of animals, fish and milk and wage employment. Income in Ihanda derives from coffee, maize, beans, groundnuts and vegetables (particularly tomatoes and spinach). Besides farming, the villagers earn money through the sale of animals, petty trading and salaried employment. Fish farming presently provides the farmers with a source of relish; very little money is earned from this activity.

3.5 Integration of farm activities

Many fish farmers in both Ihanda and Subira keep a few animals and grow a variety of other crops. The better managing fish farmers are aware that various activities can assist each other in order to increase the yield. In both villages respondents use animal manure to enrich the water with natural foods. Manure is also used in the maize and vegetable farms. Vegetable leaves from cassava, sweet potatoes, yam, cabbage and lettuce are used as feeds in fish farming. Beans are used as by-product in the maize farm and maize bran as feed in fish farming as well as in animal husbandry.

3.6 Access to and use of inputs for semi-intensive fish farming

In Subira, fingerlings are usually retained by the farmer, though according to Seki and Maly (1993) this is a less common practice in the region as a whole. Occasionally, fish farmers buy fingerlings in the village at the price of T.shs. 5 per fingerling. A number of fish farmers are known to have a surplus to provide for other farmers. Fingerlings can also be bought at the Government ponds at Luhira, 8 km out of Songea. Fish farmers acquire fingerlings only rarely through this source as there is a chance that many fingerlings would not survive such a journey on the back of a bicycle, the most common means of transport. Besides, some fish farmers complain that the fingerlings from Luhira tend to stunt. The species grown are not known to the fish farmers but according to District Fisheries Officer the most commonly produced fish is Oreochromis niloticus

Feeds, which include maize and rice bran, vegetable leaves, kitchen wastes, local brew leftovers, and manure (cow, goat and chicken), are available to most farmers in some parts of the year. However, in the rainy season, the bran is limited and manure becomes difficult to collect. Hence it has to be purchased from mills or obtained from neighbours (see Diagrams 1 and 2 on page 15). Manure can be easily obtained only from farmers who keep their livestock in a stable. Unavailability of these inputs and lack of time are for some farmers a reason for not feeding and manuring more frequently. For the most intensively managing fish farmers feeding is done twice per week and manuring once or twice per month. Inorganic fertilizer was used until a few years ago, but rapidly increasing prices inhibit this practice today.

Yields obtained from the semi-intensively managed ponds were estimated to be 1.3 – 2.0 t/ha/year, based on figures provided by the respondents. Yields are likely to be higher than this as fish farmers also harvest intermittently but often focus on the amount harvested at the batch harvest. According to Seki and Maly (1993), about 36% of fish from the ponds in the region are obtained through intermittent harvest. Hence, accurate figures on the harvest are difficult to obtain. One fish farmer claimed to obtain a harvest equivalent to 5 t/ha/year. Though he is perceived as one of the more successful fish farmers in the village the two ponds had clear water at the time of the study. The average yield from Songea Urban District is 1.1 t/ha/year while the figure for Ruvuma Region is 1.3 t/ha/year¹². According to Seki and Maly (1993) productivity is inversely related to pond size because of the ability to manage smaller ponds better.

¹² Seki and Maly (1993).

Fish farmers often know how they can improve the yield: by feeding and manuring more frequently, by maintaining the pond better, by getting new fingerlings, and by hindering otters and other predators access to the pond site. The perceived constraints are lack of capital and labour to attend these activities.

In Ihanda, fingerlings are most often obtained from other fish farmers in the village when not retained by the farmers themselves. Msuya (1992), however, mentions that individual fish farmers in the region do not keep the fingerlings. Some farmers have previously provided fingerlings free of charge, but the trend today is that they charge T.shs. 5–10 per fingerling. The tarmac road leading to Vwawa where there is a District Fisheries Office, makes it feasible for some fish farmers to buy them from the Government ponds where the price is T.shs. 5 per fingerling. But farmers still complain about the unavailability of fingerlings.

The primary source of feeds (maize bran, grass, vegetable leaves and local brew leftovers) and manure (cow and goat dung) is the farmers' own farm by-products. In times of scarcity these inputs are obtained from neighbours (often for free when in smaller amounts) or purchased at T.shs. 400–500 per bag of maize bran and T.shs. 600 per bag of manure¹³. Some fish farmers used to apply inorganic fertilizer, but as prices have escalated over the past few years it is only cash crops which benefit from this precious additive today. Feeding is done 2–3 times per week and for one respondent, twice per day. In the latter case, a farming cooperative operates three ponds and a boy is hired to take care of the daily management. The better-managed ponds in the village are fertilized once or twice per week, subject to the availability of animal manure (It is particularly difficult to collect during the rainy season.)

Yield estimates based on figures obtained from the respondents range between 0.7 and 4.0 t/ha/year. The average yield for Mbeya Region is estimated to be 0.8 t/ha/year¹⁴. The fish farmers suggest that their yields could be higher if they fed and manured more frequently. Farmers believe that many fingerlings die when ponds are harvested by the process of drainage.

Compared to the average of fish farmers in the two regions, the semi-intensive fish farmers in Subira and Ihanda manage their fish ponds better, as they all feed their fish, and all except one fertilize their ponds. Figures reveal that this is done by only 44% in Ruvuma Region and 57% in Mbeya Region (Table 3). Feeding is typically done 2–3 times per week in Ruvuma and once per week in Mbeya Region. Fertilizing is most often done on a monthly basis in Mbeya Region¹⁵.

¹³ One bag is equivalent to 120 litres.
¹⁴ Msuya (1992).
¹⁵ Seki and Maly (1993) and Msuya (1992). Data on frequency of fertilizing for Ruvuma Region is not available.

Table 3. Management practices in Ruvuma and Mbeya Regions.

Source: Seki and Maly (1993, data extracted from the socio-economic survey, Ruvuma Region) and Msuya (1992).

3.7 Activities competing with fish farming

Competition for Land

In Subira, the areas close to the rivers and streams are used for fish farming and vegetable gardening. But as unutilized land suitable for these activities is still available, and the farmer can usually borrow more land if needed, the competition between these two crops in terms of land is not severe. Land suitable for fish farming and vegetable gardening in Ihanda is entirely occupied. Consequently, these activities compete for land. However, the fish farmers interviewed indicated that they could construct more ponds by sacrificing land at the vegetable garden, if they so desired. Although vegetables are a source of income, respondents say they prefer to devote some of the land to fish farming as it is a cheap source of relish. No other activities were found competing with fish farming for land.

Competition for labour

In Subira, pond construction and to a lesser extent pond repair is usually carried out by hired labour. As a person experienced in pond construction is available in the village, the fish farmers find it easier to hire him. Since he is expensive, one wonders why the fish farmers do not try to construct the ponds themselves. All other activities are performed by members of the household. In Ruvuma, 75% of the fish farmers have constructed their pond using only household labour. There is no typical seasonality of labour in fish farming, except for pond construction and maintenance which have to be done in the months following the rainy season. Hence, all other activities are done at any time of the year. Lack of labour power is considered to be a constraint as fish farming competes with the more labour-intensive farm activities. However, the constraint was considered to be less important in Subira than in Ihanda.

In Ihanda, all fish farming activities are carried out by members of the household. For pond construction, however, the farmer is often assisted by hired labour. Because the task is laborious and time consuming, fish farmers usually construct ponds in the slack months after the rainy season, i.e. May to August. Cleaning of pond slopes is done three times every year and harvesting usually twice per year, while all other fish farming activities are carried out any time of the year. High demand for labour for other activities during peak months makes it difficult to manage the ponds properly. This is particularly the case in those families where only one member of the household is responsible for pond management. During labour-scarce months, feeding and manuring are the activities most affected.

Competition for Capital

The initial capital requirement for fish farming in Subira is high because of farmers' preference for hiring labour for pond construction and pond repair. Other capital costs include feeding and manuring when the farmers are not able to provide these inputs from their own source. Although capital is a limitation in parts of the year, the relatively low priority given to fish farming is likely to be another factor leading to extensive management. In Ihanda, apart from the initial investment for acquiring new land and hiring labour for pond construction, capital for fish farming activities is generally not a constraint, as many farmers earn income from crop sales. However, the relatively low priority given to fish farming by many farmers often influences pond management.

Competition for feeds and fertilizer

Vegetable gardening and maize farming compete with fish farming for manure while animal husbandry is a competitor for maize bran. Most fish farmers in both villages apply these inputs to all their farm activities though the amount applied reflects their knowledge on requirements and the priority they give to the various activities. Inorganic fertilizer is no longer applied to the ponds, as it has become increasingly expensive, and is therefore applied only to crops like maize for which it is imperative.

3.8 Institutions and persons important for fish farming

Respondents and PRA participants in Subira and Ihanda have obtained knowledge about fish farming from other fish farmers, US Peace Corps Volunteers (PCV) and various visiting researchers. Only a few indicated that they have obtained knowledge from District Fisheries Offices. Of these extension agents, fellow fish farmers tend to be the most important source (see Venn Diagrams 1 and 2 below), although the information provided, according to some respondents, may not always be accurate. Findings from the present study reveal that many fish farmers, although possessing the basic skills, still lack knowledge and would like to learn more.

In Subira, farmers have further received advice from a PCV based at the District Fisheries Office and from researchers (Seki and Maly's survey team). Two PCVs were based in Ihanda, one in 1985–87 and another in 1988–89. From 1989 to 1991 a Fisheries Extension Agent was based in the village. He is now based in the District Fisheries Office and claims to be visiting fish farmers in Ihanda once per week. Many respondents said they had not been visited by these extension agents.

According to Fisheries Officers in Ruvuma and Mbeya Regions, no special attention is paid to women participating in or wanting to start fish farming; extension advice is provided to every one who is interested.

As Venn Diagram 1 for Subira shows, other institutions are also important for the fish farmers. The maize mill (3) sells maize bran; various shops in Songea (5) supply pipes and nets, and the market in the village or town (6) serves as an outlet for the fish harvested. Besides providing extension services to a few farmers, the District Fisheries Office (4) lets out a fishing net to fish farmers. Politicians (7) contribute to fish farming by encouraging farmers to undertake the activity in order to raise their incomes and alleviate malnutrition.

Venn Diagram 1 on institutions and persons important for fish farmers in Subira Village.

Note: The institutions and person are ranked in order of importance, the most important being no. 1.

Venn Diagram 2 on institutions and persons important for fish farmers in Ihanda Village.

Note: The institutions and person are ranked in order of importance, the most important being no. 1.

In Ihanda (Venn Diagram 2) suppliers of inputs for fish farming -- maize bran (1), vegetable leaves (2) and manure (3) -- are the most important institutions for fish farmers. The contributions of DFO (6), the market (8) and shops (9) are the same as in Subira Village. A church (10) has organized an extension seminar in Mbeya Town on agricultural and rural development and fish farming was one of the subjects. The participants have then passed the knowledge on to other church members in the village.

3.9 Attributes of fish farming

3.9.1 Social and cultural values

Fish farming can be impeded or encouraged by social or cultural beliefs related to either the practice of fish farming or the consumption of the product.

Taboos and religious beliefs

There are no taboos (e.g. for handling manure) connected to the production side of fish farming in Subira or Ihanda. Some villagers are restricted by their religion from eating unscaled fish, like catfish, and even to use utensils that have been in contact with these fish. Previously, it used to be believed in Ihanda that eating fish would make children lose their hair.

Jealousy

Success in fish farming apparently sparks jealousy, though respondents who mention this add that it is not a major problem. Victims of such jealousy have had their ponds poisoned and their fish stolen, especially during the night and the rainy season when pond owners tend to relax their guard. However, theft of fish could result from simple economic need, the need for relish, rather than from jealousy.

Witchcraft is commonly deployed against fellow villagers, either because of jealousy or unsettled disputes. Subira supposedly has 5–10 witch doctors or “bearers of bad spirit”. They can be paid to send a predator to a villager's pond. Belief has it that to counter this attack, the villager must press into service another witch doctor who will prevent further predation. It is believed that otters which are not caught by set traps are sent by a witch doctor. In Ihanda, a key informant said that jealousy could trigger hostile action against progressive villagers, and even cause prolonged illness.

Two key informants in Subira said that success in fish farming creates envy; the envy can also lead to imitation of the successful activity. One fish farmer in Subira suffered social exclusion because of his success, but said it was not a major problem. The most successful fish farmer in Ihanda (the farm co-operative) said that jealousy was understandable because of the success of the farm. But such feelings were tempered by the fact that the co-operative provided as many as 20 to 30 jobs throughout the farming season.

Status

Farm activities are a source of income; possession of major farm assets tends to confer high status on the owners. It is therefore not surprising that, in Subira, livestock keeping is associated with the highest status (Table 4A). It is a sign of wealth and provides constant income through sale of milk.

Table 4.A. Status ranking of farm activities as perceived by respondents in Subira Village.

Fish and maize farming are also ranked by farmers as high-status activities. (Ownership of a pond confers status by itself besides being a source of relish; maize farming is associated with status because it is a staple food, i.e. a source of life, but also a source of income). One fish farmer remarked that the more and bigger ponds a villager owns, the higher the status he attains. In this case his wealth is more visible as well -- another important status factor.

Farmers in Ihanda grow a number of crops which are primarily produced for the market (coffee, groundnuts, beans, and to some extent maize and vegetables). These income-generating activities do not reflect the highest status. (Table 4 B).

Among the respondents, fish farming and livestock-keeping enjoy the highest status, despite the limited nature of these activities in the village. This is because the fish farmers interviewed are pond-owners, and to some extent cattle-owners. The reason given for the high status of fish farming is that it is a source of relish; one respondent considered it an exclusive activity -- few farmers own ponds. Another fish farmer emphasized that he was often visited by guests and experts because of his fish farming activity.

Coffee-growing was at one time considered a high-status activity; but it enjoys only a secondary position now, possibly because of the widespread Coffee Berry Disease which has affected many farms, also because of delays in payment (of one or two years) for coffee by the Regional Cooperative Union.

A word of caution: the answers concerning the status of various activities may have been biased in favour of fish farming, as the respondents were aware of the background of the interviewers. Fish farming may have been ranked high in order to please the interviewers or to obtain better extension services or development assistance in the future. This bias has to be taken into account.

Table 4.B. Status ranking of farm activities as perceived by respondents in Ihanda Village.

¹⁾ Fish farming was not mentioned as a “status activity” by this respondent.

Table 5.A. Ranking of profitability of farm activities among respondents in Subira Village.

¹⁾ Farmer no. 6 grows fish only for home consumption.

Attitude towards improved technologies

In Subira as well as in Ihanda, respondents generally have a positive attitude to improved technologies so long as they promise a higher yield. Whether they would actually implement such an appropriate new technology will of course depend on a number of factors, i.e. the cost involved, availability of inputs, labour time saved, expected returns and appropriate extension service received.

It is believed that fellow-villagers are more likely to take up fish farming if they observe pond harvesting and other pond activities. But lack of further support and knowledge as well as lack of land and capital are inhibiting factors.

3.9.2 Felt needs for adoption of fish farming

In Subira as well as in Ihanda, the respondents felt the need to undertake fish farming -- to obtain fish as a source of relish as well as income. This need has been met only partially in Subira as the people there do not eat fish as frequently as they would like to, and income earned from this activity is not as high as they had expected. It is worthwhile pointing out here that a few respondents have managed to establish other activities or to construct houses using cash obtained partly through fish farming.

The need for fish as a source of relish has been met partially by fish farmers in Ihanda, but only a few fish farmers earn cash from the activity. Despite failure to fully meet the needs, most fish farmers in the two villages have continued with fish farming, hoping that it will be more successful in the future.

3.9.3 Complexity

All respondents in Subira and Ihanda regarded pond construction as the most difficult activity in fish farming: it was tedious and time-consuming and required hired labour. Feeding was also mentioned as a difficult activity in both villages because of the constant need for feeds. Another problem in Subira concerned the supply of feeds -- during periods of shortage they are available only in the town. In Ihanda, feeding was difficult because the pond was so far from the farmer's house. Poor roads and non-availability of transport facilities meant that farmers have to walk or use bicycles to transport these inputs. Another activity considered difficult in Ihanda is pond draining and cleaning.

3.9.4 Relative advantage of fish farming compared to other farm activities

Profitability

Fish farming is ranked high compared to other farm activities in terms of profitability by the respondents in Subira (Table 5A). Other profitable activities include maize farming and vegetable gardening. In Ihanda maize farming ranks as the highest activity, typically followed by vegetable gardening, beans and coffee farming (Table 5B). Fish farming is perceived here as a low-profit activity as most fish harvested are consumed within the household.

Table 5.B. Ranking of profitability of farm activities among respondents in Ihanda Village.

¹⁾ Farmers no. 2, 4 and 6 only grow fish for home consumption.

Marketability

In both villages fish was considered easier to sell than other products (Tables 6A and 6B). Most respondents tend to sell fish at their pond sites and the remaining amount is usually disposed of in the village market (see Venn Diagrams 1 and 2 on page 15). The more intensively managing fish farmers among the respondents in Subira indicated that they had sold fish at the market in Songea. In both villages the supply of fish does not keep pace with demand. Other products that are easy to sell include maize, beans and vegetables. Coffee and groundnuts were considered more difficult to sell. (Coffee, because of recent problems with low prices and delays in payment.)

Table 6.A. Ranking of marketability of crops among respondents in Subira Village.

¹⁾ Farmer no. 5: the head of household and his son -- the actual farm operator -- disagreed on the ranking.
²⁾ Farmer no. 6 grows fish only for home consumption.

Table 6.B. Ranking of marketability of crops among respondents in Ihanda Village.

¹⁾ Farmers no. 2, 4 and 6 grow fish only for home consumption.

Source of relish

In both villages, fresh fish was the most preferred source of relish among respondents (Tables 7A and 7B). This was followed by chicken, cow and goat. Many respondents indicated that fish is preferred not only because it is more palatable but also because it can be obtained in smaller units than other sources of relish. They said that while it is possible to buy a small fish at a low price [T.shs. 60],

it is not easy to get beef, goat or chicken at that price. Similarly, it is easy to obtain fish as a source of relish through intermittent harvesting, but it is more costly to slaughter a cow, a goat or a chicken when wanted or when guests come calling.

Table 7.A. Ranking of preference of relish among respondents in Subira Village.

¹⁾ Farmer no. 1: the wife and the husband (both pond owners) disagreed on the ranking.
²⁾ Farmer no. 5: the head of household and his son -- the actual farm operator -- disagreed on the ranking.

Table 7.B. Ranking of preference of relish among respondents in Ihanda Village.

Initial cost

In both Subira and Ihanda, maize farming was said to generate higher initial costs than other activities (Tables 8A and 8B). Maize farming requires hired labour for land clearing and tilling, purchase of fertilizer and seeds. Fish farming was ranked second in Subira because respondents prefer to hire labour for pond construction. In Ihanda fish farming was ranked low, because ponds are constructed by the fish farmer himself though often with assistance from hired labour.

Table 8.A. Ranking of initial costs for farm activities among respondents in Subira Village.

Table 8.B. Ranking of initial costs for farm activities among respondents in Ihanda Village.

Immediacy of reward

In Subira as well as in Ihanda, vegetable gardening and beans farming offered more immediate rewards (within two to three months) than other activities (Tables 9A & 9B). In Subira fish farming ranked next, as many farmers harvest after six months. The few respondents who harvest after a year are also less intensive in the management of their pond. Maize and coffee offer their rewards after seven and 36 months respectively. Half of the respondents in Ihanda harvest after one year, thus making fish farming a late-reward activity. However, some others manage to harvest after three to six months. Other late-reward activities include maize (seven months), groundnuts (eight months) and coffee farming (36 months).

Risks

Most respondents in Subira and Ihanda say that maize farming is riskier than other activities (Tables 10A & 10B). Risk factors in maize farming are failure to purchase fertilizer because of unavailability or high price, too much rain and wind. In both villages, fish farming was seen as a relatively low-risk activity. The risk elements include too much rain -- which may flood the pond --and predation by otters. Other elements like theft, poisoning and lack of water were less important.

Table 9.A. Ranking of immediacy of reward of farm activities among respondents in Subira Village.