Provinces, included in the study are indicated by a pin. These are:
The study covered a total of 17 districts.
N.B In Nusa Tenggara province the islands of Lombok and Sumbawa were both included.



During 1995 the Ministry of Agriculture of Indonesia conducted a study on soybean seed systems in Indonesia.

The objective of the study of “Local Soybean Seed Systems “ in Indonesia was to find out how traditional farmers in Indonesia manage to obtain viable soybean seed, while planting only one soybean crop per year, purchasing only 13% of the total seed requirements as certified seed and given that soybean seed looses its viability rapidly in less than three months after harvesting.   

The study covered the five Indonesian Provinces with the largest area of soybean cultivation, covering 900 000 ha of the total area of 1 500 000ha planted with soybean in in 1994/1995 in the provinces of Aceh, North Sumatra, West Java, East Java and Nusa Tenggara Barat- Lombok & Sumbawa.  A team of two senior and five junior Indonesian agricultural researchers under my directorship completed the survey for this study during May- July 1995.

A summary of the report, written by the two senior Indonesian researchers and myself is presented here as a companion to images 198-261 of the Indonesian Photo Gallery of my professional work.

The importance of soybean in Indonesian food crop agriculture:
Soybean, after rice and maize is the third most important food crop in Indonesia. The genetic center or origin of soybean is located in North China. However soybean has been grown on Java since hundreds of years and was probably introduced by Chinese immigrants before 1600. Since the 1950’s soybean production has rapidly spread to the outer islands, however, demand increased even faster resulting in large soybean imports to cover the shortfall.  In November 1985 the Government of Indonesia therefore launched a special soybean development program to again achieve self - sufficiency in soybeans existing before the seventies. This effort resulted in an increase in the annually planted area with soybean from 900 000 ha in 1984 to over
1 500 000 ha in 1994, while average yield levels increased from 900 kg per ha in 1984 to 1 100 kg per ha in 1994. Yet only 13 % of the seed in 1994 was purchased from Government and commercial seed companies. This implies that 65 000 tons of soybean seed or 87% of seed was handled by local seed systems, either on-farm saved, locally purchased or exchanged.

Table 1 Planted area of Soybean in the five Provinces of Indonesia with the largest area of soybean in 1994


No of ha planted in 1994



North Sumatra


West Java


East Java


Nusa Tenggara Barat




Soybean as a food crop
Soybean is an important food crop with high protein contents of about 40% and fat contents of 20%. The annual consumption per person in Indonesia is 16 kg, of which 60% for food and 40% for livestock feed. Soybean is not directly consumed but is processed into products such as tahu (coagulated bean curd): tempe (fermented soybean cake) and ketjap (soybean sauce) or used for livestock feed or industrial purposes.

Farmer’s seed supply systems.
Seed is an indispensable input for agriculture and farmers obtain their seed from the following sources:•    Grain retained on-farm and used as seed

  • Grain bartered at the local level and used as seed
  • Grain sold as unlabeled seed brought from the local market
  • Certified or labeled seed bought from the formal distribution system

Reviewing the main food crops in Indonesia as grown by traditional smallholder farmers, it is known that the bulk of the seed for food crops is supplied by local seed systems, which are based on on-farm saved seed or seed obtained locally from other farmers or traders.

Supply of rice and maize seed:
Rice and maize seed in traditional agriculture is retained on-farm grain, while in development programs farmers receive improved rice seeds from the development agency concerned, or buy improved seeds from the seed rice industry. It is furthermore an established fact that seed from traditional rice and maize varieties and landraces can be stored on farm for several seasons, while maintaining a high viability.

Supply of soybean seed:
 Soybean seeds rapidly loose viability and can only be stored for periods of less than three months. However, most traditional smallholder farmers in Indonesia who plant soybean, plant only one soybean crop per agricultural year, mainly as a secondary crop after rice or maize. The question arises how, considering these conditions, can they obtain viable soybean seed?

The main purpose of the study
The main purpose of the study was to find out how the Indonesian farmers who plant soybean obtain their seed, given the situation that most farmers plant only one soybean crop per year and that the viability of soybean seeds is less than three months. The total soybean area cultivated in Indonesia in 1994 was about 1.5 million ha with an average seed rate of 50 kg per hectare and therefore the total amount of soybean seed needed per year is about 75,000 ton per year. In the season 1994/95 only 10,000 tons of soybean were provided by Government agencies and commercial companies, which implies that an amount of 65,000 tons of soybean seeds were locally produced or exchanged.

Up to 1995 only a few studies of the soybean industry had been made, of which the Lenggogeni report of 1990 is the most complete study. The main focus of the Lennogeni report was at the macro level, identifying seed flows major production areas to major consumer areas of soybean in Indonesia, while only limited attention was given to the seed flow between farmers. This study mentions however, a specific local seed supply system, with a seed flow between farmers called “Inter-field seed flows” or JABAL system (Jalur benih antar lapang)

The purpose of the present study was therefore to find out in detail:

  • Farmers’ source and management of soybean seed
  • The role of soybean in the farming system
  • The local soybean and grain market

And in addition to collect primary information on:

  • Costs and returns of soybean production
  • Local soybean seed growers
  • Farmers ‘soybean seed: quality; identification of varieties grown by farmers; 
  • Yields obtained with farmers’ soybean seed.

To conduct the fieldwork a team of seven Indonesian agricultural researchers under my guidance visited the provinces of East-Java, Nusa Tenggara Barat, Aceh, North Sumatra and West Java during May –July 1995.

The team interviewed 510 soybean- growing farmers, 69 traders and six seed growers. Also 87 district reports and 89 seed samples during harvesting or planting soybean were collected.

Description of the JABAL system
The main finding of the study was a detailed description of the “JABAL” system or “Inter field seed flow system for soybean”.

It was confirmed that in each major soybean producing area, the agricultural year consists of three seasons and that in each season soybean are planted in a particular land use type. To illustrate the “JABAL” system, a description is given of the soybean seed supply in Lamongan district of East Java Province. The map of this district illustrates clearly the flow of soybean seed between the six sub districts of Lamongan, using the three planting seasons followed in this district: (Image 198, 217 & 223)
-On dryland in Kecematan Mantup during the rainy season
- On rainfed sawahs during the first dry season in Kecematan Kembangbahu and Tikung .
-On irrigated sawahs during the second dry season in Kecematan Sukodadi, Sugio and Kedungpring. See following map:

Planting season

Land use type

Farmers type of operation

Rainy season:

November – February

Dry land:

Kecematan Mantup

Farmers operating dry lands

First dry season

February –May

Rain fed sawahs:

Kecematan Kembangbahu &Tikung

Farmers operating rainfed sawahs

Second dry season

June- September

Irrigated sawahs:

Kecematan Sukodadi, Sugio and Kedungpring

Farmers operating irrigated sawahs

  1. Farmers operating dry lands, that is higher located lands, often on the slopes of mountains, plant soybean in the rainy season, from November to February, buying seeds from farmers who planted soybean during the previous second dry season, some directly but mainly through local village and district soybean dealers.
  2. Farmers operating rain fed sawahs plant soybean during the first dry season from February to May, buying seeds from farmers who planted soybean during the previous rainy dry season, some directly but mainly through local village and district soybean dealers.
  3. Farmers operating irrigated sawahs plant soybean during the second dry season from June to September, buying seeds from farmers who planted soybean during the previous first dry season, some directly but mainly through local village and district soybean dealers.
  4. After the above agricultural year the cycle repeats it self and farmers operating dry lands will plant again soybean in the period from November to February obtaining seeds from the farmers, who planted soybean during the last years’ second dry season on irrigated sawahs, thus continuing the cycle.


Source: Local Soybean Seed Systems in Indonesia 1996:
van Santen C.E, Heriyanto & Eko Legowo


Summary description of the JABEL system

The JABAL system or the “Inter field soybean seed supply system” operates in the study area of the five provinces and other soybean production areas of Indonesia. For the year 1995 the JABAL system supplied 66% of all soybean farms with seed, while on-farm saved seed is reported in 34% of the soybean farms.

The operation of the JABAL system is based on the combination of three features:

  • The diversity of physical conditions found in all major soybean producing areas in Indonesia, which results in three land use types: dry lands, wetlands rain fed and wetlands irrigated, which in most areas are located within a short distance of each other.
  • Due to the relative high rainfall with favorable annual distribution, sufficient water is available in food crop producing areas in Indonesia in most years to allow farmers to grow two to three successive crops, among which soybean. Crop seasons are: rainy season, 1st dry season and 2nd dry season.
  • The fact that soybean genotypes to date planted in Indonesia appear suitable for rainy season as well as for the first and second dry season plantings  (Sumarno, 1985)

Soybean is thus grown during three cropping seasons, using one of three land use types. This allows the movement of seed between seasons and land use types, thereby avoiding the problem of reduced germination due to seed storage over one or more cropping seasons.

Thus, for example, seed needed for dry land plantings during the rainy season is ob¬tained from irrigated wetlands where soybean was planted during the second dry season from the previous crop year. Farmers planting soybean in wetland-rain fed during the first dry season obtain their soybean seed from the dry land soybean crop grown during the previous raining season, and in their turn supply seed for the wetland-irrigated soybean crop, which is planted during the second dry season, after which the cycle is repeated.   

Some farmers in E. Java, who operate all three land use types, will save soybean seed from the previous crop as planting material for the next crop. Other farmers, using the same land use type for three successive crops of soybean in one year, as for example on dry lands in Aceh Province, will also use on-farm saved seed.

Other farmers, who can only grow one soybean crop per year, plant soybean during the off-season on small plots on the bunds or in a corner of their fields to save seed for the next crop. In other areas, where farmers plant two successive soybean crops per year, they will buy seed for the first crop and use on-farm saved seed for the second soybean crop. Some farmers in W. Java Province, have found a method to store soybean seed in air tight tins, mixed with ash. In this way they are able to keep on-farm saved seed for up to nine months with hardly any reduction in germination and vigor.

However, the majority of the farmers (66%) report that they buy seed, at least once per year. This applies in particular to farmers, who plant one soybean crop per year and to farmers, who plant two crops per year, for their first crop. 

Farmers, who buy seed, indicate that the most important criteria for them of the seed to be purchased, is that this seed has been recently harvested. Thus, when buying seed from other farmers or traders, the farmer makes sure that the seed has been recently harvested, guaranteeing that germination and viability will be sufficient.

The team found that in East Java and NTB basically all seed purchased is supplied from within the same district or from an adjacent district.  This is feasible as all three land-use types are present and all three crop-seasons are used for soybean production in all districts in East Java. Obviously, farmers living in a specific sub-district will purchase seed from an adjacent sub-district of another district, rather than from a far away sub-district. The survey consistent¬ly found that in all cases when seed was bought, this was obtained from within a distance of at most 30 kilometers.

Another feature is that in most districts, an overlap in time exists between the harvest period of the previous soybean crop and the planting period for the following soybean crop.  Actually the survey team observed in a number of locations in adjacent fields some farmers harvest¬ing, others threshing and again other neighbors already planting soybean, all on the same day. In most areas farmers report that usually soybean seed is on average planted within one week after being harvested.

For all other districts in the five provinces surveyed, similar maps can be drawn as map on page 17, with local seed flows between seasons and land use types. According to staff from the Provincial Agricultural Services, from the five provinces, similar seed flows exist in the other districts, not included in the present survey.

In summary, the JABAL system represents a low cost and efficient soybean seed supply system, operated by farmers and local traders, established by local initiative, without Government assistance. The existence of this local seed system explains why there is little or no urge for farmers to buy certified soybean seed from the formal seed system, especially when it cannot be proven that this certified material performs superior, as is the case to date.  The best and most efficient development strategy for the Government would therefore appear not to intervene in the JABAL system, as long as no superior new soybean varieties are available. Once, when new improved soybean varieties become available, these could be most efficiently introduced using the JABAL system, rather than creating an additional seed distribution system. The introduction of these new varieties could be best done through a combination of demonstration plots and farm days. Distributing on these farm days small quantities of this improved material to interested farmers ensures that these new varieties spontaneously enter the JABAL system, provided that farmers are convinced of the superior performance of these new varieties.



previous image | next image