English (en) | Change Language
By: Darrell Cox
Published: 2002-04-20


An article in the October 1997 ILEIA _Newsletter_ tells the story of a land development that was established in the 1980s in Sri Lanka. Farmers were provided with one hectare of irrigable rice land that yielded five to six tons of rice without fertilizer during the initial three to four years of monocrop culture. After that, yield declines were experienced that compelled farmers to apply inorganic fertilizers to maintain yields at four and one half to five tons per hectare. The government removed subsidies on fertilizer, and the profitability of rice production declined substantially.

“One of the important reasons for yield decline in irrigated rice is nutrient mining. Total nutrient removal per hectare by a rice crop of five tons amounts to about 100 kg of nitrogen (N), 16 kg of phosphorus (P) and 128 kg of potassium (K) [220, 35, and 282 lbs. of N, P, and K, respectively]. Farmers compensate for this loss by applying about 375 kg [827 lbs.] of inorganic fertilizers which contribute 117 kg of N, 23 kg of P and 42 kg of K (258, 51, and 93 lbs. of N, P, and K, respectively).”

Researchers and extension workers in Sri Lanka began working with farmers to develop methods to maintain soil fertility. One result was that farmers began incorporating rice straw back into the soil rather than burning it during the threshing process, which was the traditional method.

Secondly, a researcher in the Sri Lankan national agricultural research system began screening a number of green manure crops for biomass production and N-content. Sesbania rostrata stood out among the species that were screened (others included S. sesban, S. aculeata, Crotalaria juncea and C. caricia). Sesbania rostrata is an annual Sesbania species; it is an aquatic legume that can be grown before or between rice or maize crops as a green manure. S. rostrata is the only species of Sesbania that has nitrogen-fixing nodules on its stems as well as its roots.

The research results were remarkable. S. rostrata, planted at a rate of 60 plants per square meter, was able to produce 4000 kg of dry matter and 100 kg of N per hectare (1.8 tons dry matter and 89 lbs. of N per acre) after just 45 days of growth.

Three traits make S. rostrata suitable for a paddy rice / green manure production system: 1) its rapid growth and substantial biomass production in a short period of time, 2) its superior ability to fix substantial amounts of nitrogen, and 3) its ability to grow well (and fix nitrogen in stem nodules) in water-logged soil. As a result, S. rostrata has the potential to fit into the fallow period between rice crops in systems where two rice crops are grown in a single year in the same paddy.

When S. rostrata is sown for the first time in a field, the seed needs to be inoculated with a Rhizobium bacterium to ensure optimal nitrogen fixation. The following seed treatment was recommended in a 1991 issue of Entre-Nous (a newsletter for French-speaking Africa). The seed should be scarified with boiling water by pouring water over the seed and letting the seed soak overnight. (note: Sesbania is hardseeded and farmers must overseed their land to obtain a desired plant population. Other treatments used to break dormancy include sand scarification and treatment with concentrated sulfuric acid.)

The inoculant is prepared in the following way: put the inoculant in a plastic bag, add water, close the bag and mix well, and then leave it in a cool place overnight. After the seed has been drained (following hot water treatment), it can be mixed with the inoculant. Again, if Sesbania is being planted for the first time, it is advisable also to inoculate the stems when they reach a height of about 25-30 cm (10-12 inches). The inoculant is prepared in the same way as explained above. The next day it is diluted with one liter of water and sprayed on plants using an ordinary sprayer.

I contacted AgroForester (a company that sells inoculant) to find out more about the rhizobia that is used for inoculation of S. rostrata seed. One of their staff persons reported, “… as I began researching the specific species name for this rhizobia, I found that S. rostrata has root nodules of the genus Rhizobium and stem nodules of the new genus and species Azorhizobium caulinodans.” From that information he concluded that their group E inoculant should root nodulate, but may not stem nodulate S. rostrata. [Editor: After a more recent enquiry, we were told that the group E inoculant should work for both stems and roots. It contains a strain of rhizobium isolated from S. rostrata. It has been used by many for S. rostrata, and no complaints have been received by AgroForester.] The two companies listed in Tree Seed Suppliers Directory that can provide inoculant for S. rostrata are: 1) AgroForester, P.O. Box 428, Holualoa, HI 96725, USA, e-mail: seeds@agroforester.com and 2) Soil Productivity Research Laboratory, Private Bag 3757, Marondera, Zimbabwe.

Nitrogen fixation is a biological process and as such is affected by environmental conditions. Therefore, the amount of nitrogen fixed by S. rostrata will in part be determined by the growing conditions at the time. In general, nitrogen fixation can be increased by the application of P and K fertilizers when soils are low or depleted of these nutrients. However, large applications of mineral nitrogen fertilizer tend to reduce the amount of nitrogen fixation because another source of nitrogen is readily available. Recent studies on nitrogen fixation and S.rostrata confirm these results. Although P and K applications may increase the amount of nitrogen accumulation in Sesbania, rice yields do not necessarily increase, making this treatment questionable for small farmers. Soil flooding also affects biological nitrogen fixation; if soils are flooded, nitrogen fixation by nodules on the roots does not occur. Yet, because there are nodules on the stem, nitrogen fixation is still occurring above the flooded zone. It should be possible to see nodules on roots two to three weeks after germination. Functioning nodules are pink or red inside, while nonfunctioning nodules may be white, gray, or green.

The Entre-Nous article recommended that Sesbania be planted in moist, but not submerged, rice paddies six to seven weeks before rice is to be transplanted. Spacing should be 20 cm (8 inches) between seed in a row and 20 cm between rows (36 plants per square meter). When S. rostrata reaches a height of 1 to 1.5 m (40 to 60 inches), the plants can be cut and chopped into pieces less than 20 cm (8 inches) in length. The green manure should be mixed into the paddy to a depth of 10 to 15 cm (4 to 6 inches). Rice can be transplanted into the paddy one to seven days after mixing in the Sesbania.

Mike Fennema, a past ECHO intern, worked with S. rostrata as a green manure as part of a project with Food for the Hungry International (FHI) in Cambodia. I asked him to give us some perspective based on his experience. Mike said, “when I first started the project back in 1993, I talked to the staff at the International Rice Research Institute (IRRI). They had tried to promote Sesbania for several years, but had abandoned it. They had great results in terms of [rice] yield increases during the field trials, but the farmers did not seem to be interested in growing their own seed.”

Mike indicated that two things need to be taken into consideration when contemplating the use of S. rostrata as a green manure in paddy rice production: 1) growing conditions, and 2) farmer initiative, especially related to seed production. Sufficient amounts of water and heat must be present during the period between rice crops to allow S. rostrata plants to reach about one meter before being plowed into the soil. They found this much growth is necessary to provide the nutrients and organic matter that are needed for the next rice crop.

One of the reasons that IRRI gave up on using S. rostrata in rice production systems was the trouble they had in getting farmers to grow it for seed. The FHI project had much greater success because development was based on farmer-led innovation. “We decided to give Sesbania a try by starting small and involving the farmers in every step. We obtained a small amount of seed and encouraged the farmers to grow some for seed while experimenting with the rest. We helped the farmers to realize that we did not have the best techniques for growing and using Sesbania in Cambodia. They needed to help us learn about these things; when to grow Sesbania for green manure use, for seed collection, how long to let it grow before plowing it in, how tall it should be, etc. The goal was to build their confidence in learning how to experiment with a potential crop that could improve their farming system. By starting with an experiment, starting small, and involving farmers in the learning process, things improved, but the process was slow.” As of 1998 there were 1680 farmers using Sesbania rostrata as a green manure. The project is a success because enough farmers have learned to grow and harvest Sesbania for seed; excess seed has been produced that is being sold for experiments in additional villages.

ECHO now has a limited supply of Sesbania rostrata seeds. If you are working overseas in agricultural development and would like a free sample packet, please contact us.

Cite as:

Cox, D. 2002. Sesbania rostrata: A Green Manure Production System for Rice. ECHO Development Notes no. 75