The district of San Pedro is located in the Department of San Pedro in the eastern part of Paraguay. It is a relatively isolated area with poor dirt road access. Small farms in San Pedro have been cultivated in part for up to 100 years, although many areas have been cleared of forest during the last two decades. Initially soil cultivation on small farms was done manually. With the introduction of work oxen for soil cultivation in the last 20 years an expansion in the areas cultivated on small farms has occurred.

The town of San Pedro has minimal facilities. The government-owned BNF is the only commercial bank. There is also an agency of the CAH and a local co-operative (A Nortena Ycuamandyyu Ltda) provides banking and credit facilities to its rural and urban members.

A GTZ funded rural development, implemented over six years and terminated in early 1998, had a positive impact on development of the district of San Pedro. Particularly noteworthy in the context of this study was its support to the:

expansion of the Nortena Ycuamandyyu Co-operative from a tiny operation;

• establishment of a local Saturday fresh farm produce market in San Pedro;

• introduction and dissemination of mucuna;

• development of no-till on small farms particularly during the last 2 years to the project.

It is a pity the GTZ project was terminated so soon. The impression gained by the study team in its field visit to the region was that a consolidation of the project for a further 2-3 years would have resulted in a much greater impact. This comment is made with respect to the project’s support to the development and extension of no-till and green manure crops, and the small farmer co-operative. It is difficult to see any of the project’s excellent initiatives being further expanded without further external support.

Soils are sandy. Although the district is mainly of flat contour, soil erosion is a problem and causes rapid degradation of organic matter and the depletion of soil fertility. It also creates havoc with road access during heavy rainstorms.

According to the most recent census data (1991), there are almost 30,700 small farms of less than 20 ha in the Department of San Pedro occupying almost 224,400 ha. The percentages of farms of 1-5 ha, 5-10 ha and 10-20 ha were 28%, 33% and 39% respectively. Within the district of San Pedro there are about 4,200 farms, 3,000 of which are small farmers with farms of 5 to 25 ha (F. Cespedes, pers. comm.). Colonies of small farms were established in the district of San Pedro in the early to mid 1970s. Farms were mostly 18 or 20 hectares each. Families migrated to San Pedro from various parts of the country.

Cotton is the main crop grown in the district on small farms. Small farmers also rely on mandioca and maize as sources of income, and in the most recent years, through initiatives of the Co-operative, cedron and sesame have taken on some importance on a very small number of farms. Cattle, pigs, ducks and chickens also provide important sources of supplementary farm income. A few farmers sell fresh vegetables and other produce at the San Pedro Saturday market. Although important to these few farmers, and to town dwellers alike, very few benefit from this local market initiative. However, the Co-operative is well positioned to assist a significant number of small farmers to market their surplus produce, although further support will be needed in the short to medium term to realise this. With 300 urban socios and 340 rural socios the Co-operative is presently a small-scale operation reaching a very small number of small farmers in the Department of San Pedro. Its major commercial operations are cotton trading (presently 200 t per annum), purchasing maize (250 t/annum) and the preparation of feed rations 70-80 t/annum and processing of yerba mate (40-50 t/annum). Small quantities of cedron (50 t in 1998), sesame, mandioca and ca’anei are also traded by the Co-operative.

The Co-operative presently lends credit to 200 rural socios. Credit is mostly short-term to finance annual cropping. Some 60 socios have long-term loans from the Co-operative for the purchase of machinery, water supply systems, etc. According to the Co-operative, its credit operations are being impeded through the CAH offering credit at rates below what it is able to. At the time of the study visit to San Pedro, the CAH was offering credit at 17.5%/annum while the interest charged by the Co-operative was 24%/annum. The Co-operative was borrowing its funds for on-lending to farmers at 17.5%/annum (Cajas pers.comm). The CAH, which appeared to be on-lending to farmers at cost, was lending to 3,145 small farmers in the Department of San Pedro. Of these, 79% were meeting their loan obligations while the remaining 21% were behind in their repayments. Normal rates of interest charged by the Banco del Formento were from 32%-39%/annum.

An interesting new development initiative for the Co-operative is been initiated. With support of a US$500,000 loan from BID, for the construction of a new feed ration factory and yerba mate processing facility has started. This development will be very important, but more development will be urgently needed if the Co-operative is to survive.

Crop yields on small farms in San Pedro are average by Paraguayan standards. Cotton averages about 1,200 kg/ha, mandioca 20 t/ha, maiz colorada 2,000 kg/ha and maiz chipa 1,800 kg/ha. Yields have been declining because soil nutrients are lost through soil erosion in conventional cultivation, soil conservation measures are generally not undertaken and crop residues are always burned. Whenever feasible, farmers use crop rotations. Fertilisation is not practised as small farmers are not convinced of its benefit. In the case of cotton, for example, crop yields averaged 1,800 kg/ha 10 years ago and have now fallen to 1,200 kg/ha. Crop variety fertiliser input trials on farms should be an integral part of any soil conservation development and extension programme. Such a programme will need to include the combined effects of crop rotations and residual nutrients of green manure crops.

With the development of sustainable cropping systems based on no-till and green manure crops, there is no doubt that the favourable climate of San Pedro will support sustained increases in production and the incomes for small farmers. Melanite colonies of larger mechanised farmers in the Department of San Pedro have successfully adopted no-till/crop rotations and have virtually eliminated soil erosion. At the same time, quite remarkably, these farmers have increased their cropping intensities, reduced their crop production costs and lifted substantially their farm profits (Sorrenson, et al 1997). The crop rotations adopted are based on soybeans and maize (both main-season and out-of-season), and have incorporated wheat, sunflower, sorghum, as well as both winter and summer green manure crops. The warmer climate and sandy drier soils of San Pedro permit a higher cropping intensity than is achievable in the eastern/southern wetter and cooler clay soil zones of Alto Parana and Itapua.

There would seem to be no technical barriers for small farmers in San Pedro to develop similar crop rotations using no-till with animal traction. These opportunities are available. To realise them however, will require considerable Government and external support on four fronts:

• on-farm development of the systems together with small farmers on their farms;

• an extension programme to spread these programmes cost-effectively to other farmers;

• affordable long-term credit so small farmers can get access to animal traction no-till equipment;

• affordable short-term credit to finance the costs of their annual crops;

• technical assistance and credit for the small farmer co-operative.

It will also be necessary for Government to maintain accessible and affordable lines of short-term credit for small farmers through the CAH and farmer co-operatives for the planting of their crops (seeds, fertiliser, pesticides and contracted labour).

3.2.1.Introduction

A typical small farm in San Pedro today is about 10 ha and about half of the farm is cultivated each year (F. Cespedes, pers. comm.). Generally a crop rotation is practised and no soil conservation measures are undertaken. Crop residues are burnt and soils are usually ploughed 2-3 times, then disc harrowed at least once, prior to sowing a crop. Cotton is the main small farm crop. Typically a small farmer will cultivate up to 5 ha with animal traction and grow 2 ha cotton, 1 ha mandioca, 1 ha maiz, 0.5 ha of mani, banana or cedron as an additional cash crop, and 0.25-0.5 ha of various food crops for home consumption. Small farmers have fruit trees, especially citrus, banana, mango and mamon, as well as milking cattle, pigs and chickens. Most commonly there is no forest area remaining on the farm. Traditionally farmers grew tobacco on newly cleared fertile land.

The conventional farming system studied in San Pedro was that of Sr. Agustin Akino which is considered to be typical of conventionally cultivated small farms in San Pedro.

Sr. Akino started to farm his 8.5 ha property 1978. The property is located close to the main road about 5 km from San Pedro. Five hectares had been cultivated previously and the rest of the farm was covered in native forest. At first only food crops were grown and in 1981 cotton was first sown. Thereafter 1.5 – 2.5 hectares were grown each year. In 1994 mucuna was first introduced and has been grown every year since with maize. The mucuna is ploughed into the soil. In 1997 an additional 15 ha of land in natural pasture was purchased for the purpose of increasing cattle numbers.

Sr. Akino is an active community leader. He lives on his farm with his wife and four children. The eldest son helps with soil cultivation and outside labour is never contracted.

Income is earned from the sale of a variety of crop produce and animals are an important part of the farming system. Cotton, maize, mandioca and vegetables are sold from the farm plus also milk and cheese. In May 1998 one pair of work oxen, 2 horses, 20 cows (2 in lactation), 3 sows and 50 chickens were part of the farming system.

A range of hand-tools and various draft animal implements are owned by Akino. These include 2 arados, 1 rasto de disco, azadas, machetes and a bicycle.

Whenever possible, crops are rotated. Crop residues are always burnt and there are no contour banks on the farm.

The yield of cotton has been falling. Initially, cotton yielded 3,500-4,000 kg/ha but yields have now fallen to 2,500-2,700 kg/ha. Fertiliser has never been used on the farm. Sr. Akino mentioned that on his neighbour’s farm, where 3 ploughings are the norm prior to the sowing of all crops, cotton yields have fallen to 800-900 kg/ha. Mandioca has always yielded about 20 t/ha. Akino reported maize yields of 1,500-2,00 kg/ha, and although not reported, initially maize yields were probably higher than this.

The following views were expressed by Sr. Akino when he was interviewed by the study team:

• Uses conventional cultivation because of tradition and custom. He enjoys cultivating the soil and believes it is less expensive than no-till because he is able to better control weeds.

• Observes his neighbour (Sr. Ramon Oporto – see below). While he is aware of Ramon’s positive experiences with no-till, he believes that he himself cannot adopt no-till on his farm because he does not have as many parcelas as Ramon.

• While aware of the virtues of green manure crops, he only uses mucuna, which he ploughs into the soil.

The analysis is of the current conventional tillage system used. This provides a basis for comparing the financial performance of the two other neighbouring farms analysed in San Pedro where green manure crops and no-till have been introduced. The same input and product prices (May 1998 values) have been used in the analysis of this farm as are used in the analyses of the other farms. Input prices were those of May 1998 while the product prices used are average prices since these characteristically follow wide intra and inter-seasonal fluctuations. Current "expected" crop yields specific to this farm have been used.

The results of the overall farm analysis are shown below in Table 15. Detailed crop budgets are tabulated Appendix A2.1, Tables 2-7.

Cotton is the main source of farm income at US$760, just over half of the total farm net income of US$1,416. Mandioca also provides a substantial part of net farm income making up 38%. The most profitable crops are cotton and mandioca, as can be seen from the per hectare net income returns of US$380 and US$360/ha respectively. The net income from maize, even when it is grown with mucuna, is calculated to be negative.

The total labour used is 183 person-days. The main demanders of farm labour are cotton, mandioca and cattle, which combined account for 75% of total labour. The return to labour is calculated to be US$7.74/day. Compared to the rural labour rate in San Pedro of Gs10,000/day (US$3.57/day) this is quite a good return and reflects the energetic and hard-working nature of this farmer.

Looking in detail at the cotton crop budget (Appendix A2.1, Table 7), the most important costs concern soil preparation and weeding. The two ploughings cost Gs 160,000/ha (US$57/ha) and account for 14% of the total costs of growing cotton. The two ploughings are followed by one harrowing adding Gs 15,000/ha (US$5.35/ha). The combined cost of the ploughing and the harrowing are US$63/ha. The two manual and four animal traction weedings cost Gs210,000/ha (US$75/ha) some 18% of total production costs.

Considering the next most important crop, mandioca, the transport (40% of total costs) is the largest cost item (see Appendix A2.1, Table 4). However, soil preparation and weeding are the next most important cost items accounting for 12% and 19% respectively of total costs.

3.3.1.Introduction

Green manure crops were introduced into San Pedro in early 1990 through the GTZ Rural Development Project. Various species were tested, viz. vicia comum, avena negra, avena blanca, lupino azul, lupino blanco, acevem, mucuna ceniza, nabo, crotalaria and guandu. However, only mucuna has spread significantly. It is estimated that some 600 farmers now have mucuna (F. Cespedes, pers.com.). About 500 of these farmers are in the district of San Pedro and the other 100 in the neighbouring district of San Pablo. Very few of these farmers have adopted no-till and most plough mucuna into the soil. Thus the benefits of the crop protecting the soil from erosion and providing residual nutrients are only partially realised.

The adoption and spread of no-till on small farms in San Pedro is still extremely limited. It is estimated that 60 farmers last year utilised no-till on their farms and maybe as many as 300 may do this year (F. Cespedes, pers.comm). It must be emphasised that these numbers are only estimates and need to be checked in the field. What is known with certainty, is that the 20 farmers who received intensive technical support in no-till over the last 2 years from the GTZ Rural Development Project, have abandoned their ploughs and adopted no-till. These farmers are fully convinced of the value of no-till. (F. Cespedes, pers.comm.). Two of these farmers who were interviewed in-depth by the study team were already benefiting in their opinion from the adoption of no-till and green manure crops and are so enthusiastic about this. Some 25 companias received technical support from the GTZ project receiving mucuna seeds and rollo cuchillos. With an average of 15 farmers per compania (F. Cespedes, pers. comm.), the total number of farmers directly assisted has been about 375.

A number of crops are being grown under no-till on small farms in San Pedro. These include cotton, mandioca, maiz, sandia, tobacco, melon and cedron. It has been possible for farmers to consider growing tobacco again, which is a high fertility-demanding crop to be grown economically. It is interesting to note that after 1-2 years of growing mucuna in San Pedro, economic yields of tobacco are again being realised. Previously it was only possible to grow tobacco on new land after it was cleared of forest. This shows the significant residual nutrient effects of mucuna being achieved on the sandy soils of San Pedro.

3.3.2. No-till Case Study Farm I – Sr. Lucas Ledezma

The Farming System

The Ledezma farm is 5 hectares and is situated in the Compania Picada Fernandez on the main road to Nueva Germania 5 km from San Pedro township. Sr. Lucas has been farming the property for 44 years. For many years he was a socio of a local co-operative from which credit was obtained for growing cotton. Cotton was last planted on the farm in 1991. He left the co-operative three years ago and during the last two years credit has been obtained from the CAH.

The farm is worked by Lucas and his wife. Occasionally labour is hired for weeding and on rare occasions their 23 years old son helps on the farm.

Initially the farm was cultivated manually. Oxen were first used for ploughing the soil in 1978 when cotton was introduced. Cotton became the main crop up to 1991. Since then it was no longer grown and income was earned from the sale of food crops, pigs and chickens. From 1993, the opportunity was taken of selling fresh vegetables, chickens and eggs at Saturday markets in San Pedro.

The conventional cultivation system before the introduction of no-till consisted of growing about 1.5-2 ha each of mandioca and maize, as well as about 0.25 ha each of mani, poroto, sandia and melon. One pair of work oxen, 1 milking cow, some 7 pigs and 30 chickens have been a part of the system and have been maintained after the introduction of no-till two years ago.

The yields of cotton and maize were declining and tobacco could only be grown for 1 or 2 years on virgin land cleared of forest. Initially cotton yielded about 2,000 kg/ha but fell to about 1,500 kg/ha. Maize initially yielded about 3,500 kg/ha and reduced to about 2,500 kg/ha.

Mucuna was first planted in 1995. Sr. Lucas became convinced of the advantages of no-till the following year and has not used his plough since, although the work oxen have been retained for weeding purposes. Details of the evolution of no-till into the farming system are set out in Appendix A2.2. So far only mucuna has been introduced. Mucuna is planted before all of the other crops which include mandioca, maize, mani, poroto, sandia, melon and tobacco. No fertiliser is used nor are herbicides.

Sr. Lucas cited the following benefits of green manure crops and no-till on his farm:

• Protects his soil from erosion

• His soils are recuperating with increasing levels of organic matter and soil fertility

• Has lowered his requirement for labour since he does not need to prepare the soil any more

• The use of mucuna has reduced the incidences of weeds and less time is now spent weeding.

The analysis compares the financial results of the farm under conventional tillage prior the introduction no-tillage and green manure crops in 1996 to his current situation 2 years later. The same input and product prices (May 1998) are used for the two situations analysed (i.e. past and present) and "expected" crop yields are used. Those under conventional cultivation are average yields that were being realised in the latter years of conventional tillage, while those under no-tillage (present) are also averages based on Sr. Lucas’ short experience with no-till and mucuna.

The financial performance of the farm under conventional cultivation, before green manure crops and no-till were introduced, is summarised in Table 16 below.

Farm income and return to labour are low, partly explained by the small size of the farm. Total farm net income is estimated at US$571/annum. Some 164 person-days of labour were used and the return to labour was US$3.49/day. Most of the net farm income comes from mandioca, while maize is unprofitable but provides an important source of food. The contributions to net farm income of the other crops were minor in comparison to mandioca.

The most significant costs associated with the growing of mandioca and maize were those associated with soil cultivation and weeding. For mandioca, one ploughing was carried out before sowing costing Gs80,000/ha (US$29/ha) accounting for 9% of total costs, while the four weedings cost a total of Gs120,000/ha (US$43/ha) representing 14% of total costs (see Appendix A2.2, Table 6). For maize (refer to Appendix A2.2, Table 3), the two ploughings cost Gs160,000 (US$57/ha) and the three weedings Gs320,000 (US$114/ha) accounting for 21% and 42% respectively of total costs.

The financial analysis of the current situation under no-till is summarised in Table 17.

Comparing Tables 16 and 17, it is immediately obvious the major impact on farm income that no-till has had in two years, which is a very short period of time. Net farm income has risen to US$1,919 wihich is over three times that under conventional cultivation. At the same time the farm requirement for labour has fallen 6% to 154 person-days. The most dramatic changes in the farming system are:

1. maize in association with mucuna has turned maize from a negative net income in the conventional system (minus US$50/ha) to a profitable crop at US$414/ha in no-till;

2. the profitability of mandioca has increased 40% from US$327/ha to US$459/ha;

3. tobacco has been introduced, which is a highly profitable crop with an estimated per hectare net income of US$936, contributing an additional US$234 to the farm net income.

Increases in crop yields, the elimination of the costs of preparing the soil and reductions in the cost of weeding, have drastically improved the economic performance of this farming system. Yields of mandioca have increased from 20 t/ha to 25 t/ha while maize has increased from 2,500 kg/ha to 3,000 kg/ha. The costs of soil preparation have been referred to above.

The savings in the costs of weeding are substantial as illustrated below in Table 18 for the two major crops. In the case of mandioca the saving in weeding costs is US$36/ha and in maize US$64/ha. In terms of percentage of total costs, weeding reduces from 14% to 2% in mandioca and from 42% to 18% in maize.

1 Costs include the cost of weeding mucuna

Two years is a very short period of time to analyse the financial implications of no-till on this farm. However, given the small size of the farm (5 ha) and the positive attitude of the farmer to no-till, the introduction of a new cropping system has been rapid. This makes the analysis valid. It should be recognised though that there is considerable scope to introduce more crops into the farm, both cash and green manure crops, and to increase the intensity of cropping. Also it should be possible to reduce the dependence on manual and minimum-till weeding using animal traction, through the use of herbicide (especially roundup). The combined effect of these modifications is likely to further increase the financial attractiveness of no-till/green manure crops compared to conventional cropping on Sr. Lucas’ farm. 

The Oporto farm is 8.5 hectares and is neighbouring the farms of Akino and Ledezma some 5 km from San Pedro township. Sr. Oporto has been farming the property since 1972. At this time it was almost completely covered in native forest. Ramon is an active member of the local La Nortena Ycuamandyyu Co-operative.

The farm is worked by Ramon and his wife assisted by two teenage sons who are still at school.

The complement of animals which has been maintained for a number of years now consists of 1 pair of work oxen, 2 cows, 3 heifers, 1 bull, 1 horse, 4 sows and 6 fattening pigs, 3 sheep and 20 chickens.

A normal range of manual tools and draft animal equipment are owned by the Oportos. Access to a range of no-till machinery, both manual and animal traction, has been made possible through assistance from GTZ.

Initially tobacco was the main crop. One hectare was planted on new land following the clearing of forest. Over the years the area of tobacco was reduced to 0.2 ha and cotton became the main income earning crop. Cotton was first planted in 1974. Mandioca, maize and mani were grown for home consumption. Cultivation was always done with oxen.

The conventional cultivation system, before the introduction of no-till, consisted of growing about 3 hectares of cotton, 1 hectare of mandioca interplanted with maize, 0.25 ha of mani and another 0.25 ha of diverse food crops. There were 2.75 ha of native pasture, 0.75 ha of planted pasture and small areas of sugarcane and agro-forestry. These areas have been maintained after the introduction of no-till in 1996.

No specific soil conservation measures were used. Cotton was always planted in the same area and all crop residues were burnt. Soil compaction was a problem.

Cotton initially yielded an average of 3,000 kg/ha and reduced to about 2,000 kg/ha over a period of 20 years it was grown on the farm under conventional cultivation. Cotton was last grown four years ago but is planned to be grown again under no-till. Maize yields initially decreased and have maintained about 2,000 kg/ha interplanted with mandioca. Mandioca yielded about 20 t/ha. Grown with mucuna under no-till, yields of both crops have markedly increased in only two years. Grown on virgin land tobacco yielded about 2,800 kg/ha. After one or two years yields drastically reduced to 800 kg/ha or less as fertility levels rapidly declined and the crop was no longer a financially viable proposition. Under no-till, grown with mucuna, the crop is yielding 1,200 kg/ha making it again a financially viable crop to grow.

Mucuna was first planted in 1994 when 10 kg of mucuna seed was received from the GTZ project. In 1996 Ramon participated in the No-till Conference for Small Farmers in Hohenau funded by the GTZ-MAG Soil Conservation Project. From this time onwards, Ramon was convinced that no-till would provide a sustainable form of farming for him. Since this time he has never used his plough again.

So far only mucuna has been introduced into his farming system, although small areas of canavalia and guandu have been planted. There is no doubt that in the foreseeable future, with competent technical assistance, more radical changes will be made to this farming system. Mucuna has been planted with mandioca interplanted with maize. It has also been planted with mani and tobacco. To avoid problems of mucuna seeding and thereafter posing problems in subsequent crops, in future mucuna will be planted with maize for the specific purpose of producing mucuna seed. This year cotton will be replanted following mucuna. These plans have been incorporated in the financial analysis of this farming system (see following section).

No fertiliser has ever been used on this farm and there are no plans to do so. In this farmer’s opinion, fertiliser damages the soil. Herbicide has also never been used. Details of the evolution of no-till into the farming system are set out in Appendix A2.3.

In Sr. Ramon’s opinion the benefits of no-till are:

• It has eliminated soil erosion

• The fertility of his soil is recuperating

• Saves time – no need to prepare the soil for planting and less time spent weeding

• Eases work load – more time now for other activities

• Reduced reliance on cotton for income

• Sustainable form of farming

Only one inconvenience/cost was mentioned by Ramon:

• Technical assistance is essential

Full details of the financial analysis of Sr. Ramon’s farm can be found in Appendix A2.3. The analysis compares the financial results of the farm under conventional tillage before Sr. Ramon introduced no-tillage and green manure crops to his current situation two years later. The same input and product prices (May 1998) are used in the two situations analysed (i.e. past and present). In addition, "expected" crop yields are used. Those under conventional cultivation are average yields being realised in the latter years of conventional tillage, while those under no-tillage (present) are also averages based on his two-years of experience with no-till and green manure crops.

The relatively small area of food crops (about 0.25 ha) of poroto, mani and vegetables grown solely for the purpose of family consumption have been shadow-valued as maiz interplanted with mandioca.

The financial performance of the farm under conventional cultivation, before green manure crops and no-till were introduced, is summarised in Table 19 below.

Farm income and return to labour are similar to those calculated for Sr. Agustin. Total farm net income is estimated at US$1,448/annum. Some 163 person-days of labour were used and the return to labour was US$8.88/day. Most of the net farm income (56%) came from cotton, while maize and mandioca were the next most important contributing a further 29% of net farm income. In terms of returns per hectare, mandioca and maize were the most profitable followed by cotton.

The costs of soil preparation and weeding are quite a substantial part of the production costs associated with the growing of cotton and mandioca interplanted with maize. Cotton, for example, two ploughings cost Gs160,000/ha (US$57/ha) and the three weedings Gs200,000/ha (US$71/ha) accounting for 14% and 18% respectively of total costs (see Appendix A2.3, Tables 5 and 9). No-till has a major impact on these costs (see following section).

The financial analysis of the no-till farming system is summarised in Table 20.

The major impact on farm income that no-till has had on this farm is immediately obvious by comparing Table 20 to Table 19. Net farm income is estimated to have risen to US$2,538 over 75% that under conventional cultivation. At the same time the farm labour requirement has risen slightly, up 5% to 171 person-days. The return to labour has risen 67% to US$14.84/day. These results are quite amazing considering that it has only been two years since no-till has been adopted.

Using no-till in association with mucuna has considerably raised the profitability of growing cotton, mandioca/maize and mani. It has also opened up an opportunity to grow tobacco again, which is highly profitable, and to grow mucuna as a seed crop in association with maize, considerably raising the profitability of growing maize. The increases in profitability can be seen by comparing the net income per hectare figures in Tables 19 and 20. For example, the net income per hectare of cotton has increased from US$271/ha under conventional cultivation to US$458/ha in the no-till system.

Increases in crop yields, the elimination of the costs of preparing the soil and reductions in the cost of weeding, have dramatically improved the economic performance of this farming system. Yields of cotton are expected to increase from 2,000 kg/ha to at least 2,500 kg/ha. Mandioca has increased from 20 t/ha to probably much in excess of the 25 t/ha which has been included in the financial analysis, while maize has increased from 2,000 kg/ha to at least 2,500 kg/ha. The costs of soil preparation have been referred to above while the savings in the costs of weeding are significant as illustrated in Table 21. In the case of cotton, the saving in weeding costs is US$28/ha and in mandioca/maize US$20/ha. In terms of percentage of total costs, weeding reduces from 18% to 11% in cotton and from 12% to 6% in mandioca/maize.

1 Costs include the cost of weeding of mucuna

A comparison can be drawn between the economics of conventional cultivation and no-till in the three farming systems analysed in this study in San Pedro. The results are summarised in Table 22.

Despite only two years of experience with no-till on the two farms analysed in the study, the results suggest that green manure crops and no-till offers the possibility to substantially raise the profitability of small farms in San Pedro. These changes to farming systems offer a practical and economic means of raising on a sustainable basis the production and incomes of small farmers. The very low and declining incomes that these farmers were experiencing under unsustainable conventional cultivation systems, raises the importance and urgency of introducing no-till to more small farmers in San Pedro. It can be observed in Table 22 that in the two systems analysed where no-till has recently been introduced, net farm income has increased markedly. Sr. Lucas’ income has shown the most impressive rise (236%) due to the very low level of net income under the conventional system. The net income on Sr. Oporto’s farm has increased substantially by 75%. Returns to labour have all shown impressive increases.

The estimated net farm income of Sr. Agustin is very similar to that of Sr. Oporto’s before he modified his farming system by introducing no-till. This would suggest that Sr. Agustin could increase his net farm income significantly too should he substitute conventional tillage with no-till.

The analyses suggest that net incomes on small farms in San Pedro are much lower than in Edelira (compare Tables 14 and 22). This is due to fewer commercial options available to small farmers in San Pedro where infrastructure, machinery, commercial traders, farmer co-operatives and overall development are much less advanced than in Edelira. While the clay soils of Edelira are inherently more fertile than the sandy soils of San Pedro, the warmer climate and freer draining soils of San Pedro offer potential for developing more intensive cropping systems in San Pedro than in Edelira.