ceci Publication Issue n’existe pas dans votre langue, Voir dans: မြန်မာ (my), ไทย (th), ភាសាខ្មែរ (km), English (en),
Ou utilisez Google Traduction:  
Par: Jeff Rutherford

This article is from ECHO Asia Note #3

Crotalaria juncea, or sunn hemp, is a member of the pea family (Fabaceae) grown in many countries as a green manure or forage crop. Originating in South Asia, but with a common cultivar developed in Hawaii, it is a fastgrowing, drought-tolerant and aesthetically pleasing plant with real potential for integrated farming in the tropics.

Because sunn hemp is a strong nitrogen fixer with a reported resistance to root knot nematodes, and can be incorporated into the soil with little more than a month of growth, it can be used rotationally between primary crop plantings in both paddy and dryland fields. However, as sunn hemp needs well-drained soils, it does not seem appropriate for paddy farmers with drainage problems. The crop is also grown as forage and fiber, especially in South Asia. However, it is not a true hemp of the Cannabis genus.

Although Sunn hemp in northern Thailand is promoted by agencies such as the Land Development Department, it is apparently not commonly used by paddy farmers. It is known as por teuang in Thai, though it is known by some farmers as tua pui, or "fertilizer bean."

On our small experimental farm 20-km north of Chiang Mai city, we first planted a test plot, drilling a 1-cm deep hole and adding about a dozen seeds, leaving the holes uncovered. This was during the rainy season, and the sunn hemp grew rapidly and vigorously, gaining a meter of height within three weeks and flowering in less than six weeks. The bed of pretty yellow bean flowers prompted passers-by to stop and ask about them, most locals being unfamiliar with the plant. One visitor did know sunn hemp, saying that it is commonly grown in his district of Mae Sariang along the northwestern border with Burma. Farmers there grow it in their paddies between a dry-season garlic crop and wet-season rice. The man said they broadcast the crop and plow it under after a month or so, reducing their fertilizer bill.

Early production and harvesting seed

After the success of the limited trial, we began planting sunn hemp at the southern fringe of a 4-meter wide dyke separating our fish pond and the neighbor's paddy. We grew it right along the paddy to demarcate the boundary and to multiply seed. Our farmhand, possessing the local farmer's antipathy to cultivating a plant only to cut it and "throw it away," and still not clear on the concept of green manures, planted the sunn hemp in neat rows, about 30 cm apart. This is fine for seed production, but too sparse for compost material or mulch.

AN 3 Crotalaria juncea, a promising green manure crop for the tropics 1Since then, we have been self-sufficient in sunn hemp seed, harvesting the pods, which are leathery, puffy and the size of a large medicine capsule. It is easy to tell when the pods are ready to harvest: just give the plant a shake. If it makes a rattling noise, it's time to harvest (thus one common name for C rotalaria species, "rattlepod."). We strip the pods by hand, sun dry them, crush them under foot in a fertilizer sack, and winnow. It is a quite labor-intensive process, but sunn hemp seeds sell for 35 baht ($1.06 US) a kg at a nearby agricultural supply store. A program at Chiang Mai University sells seed for a more affordable 15 baht per kg, but a sack would still run more than 700 baht, making worthwhile the labor of harvesting the seed.

One neighbor approached me and told me that he saw on TV that the young flowers of sunn hemp can be eaten, most usually in an omelet. I tried it at home and it was edible, but did not make a remarkably delicious omelet. Many C rotalaria species contain toxic alkaloids, but apparently not this cultivar. Still, literature on C. juncea suggests limiting its proportion in animal feed to 45 percent or less.


After harvesting the seed, the plants quickly re-flower. We were able to get three harvests, though with decreasing yields, from the same plant. This, however, and perhaps over-production, led to a pest problem. Eventually, we expanded production to include a strip along the northern boundary, beside the canal at the eastern edge of the property, and in various places throughout the farm. This eventually led to an infestation of the larvae of what might be the blue bean butterfly (Lampides sp.). I noticed a small hole in almost every pod of older plants growing along the southern boundary, where we first began growing a lot of sunn hemp. The pods were fully formed but empty of all but dust.

AN 3 Crotalaria juncea, a promising green manure crop for the tropics 2Since then, we have tried to limit the sunn hemp seed production to widely scattered clumps around the farm. As the splash of yellow flowers is quite appealing - the dried brown seed pods much less so - sunn hemp makes an attractive addition to the garden. The flowers are very popular with bees, including the big local bumblebee. In the intervening six or so months since we noticed the infected pods, we have not observed any more pest attacks.

The dried stored seeds do not seem vulnerable to attack by pests like weevils, which are abundant in our other stores and would have ample opportunity to attack the sunn hemp seeds if so inclined.

Growing as green manure

The pest problem is only relevant for seed production. The main use of sunn hemp is to be a soil amendment whereby the plants are cut as mulch or compost or incorporated into the soil after flowering but before setting seed. According to the Tropical Forages website, this is the optimum time to make use of the crop's nitrogen-fixing capacity. "Nitrogen content is greatest at the onset of floral initiation to mid bloom, and declines as N reserves are allocated to seed development." Furthermore, at this stage the plants are not fibrous and are easy to cut and use in compost. We grow sunn hemp as a break crop and soil amender in vegetable beds.

We also tried several approaches with sunn hemp as a green manure on a larger scale, in both the paddy and a portion of land newly raised from the paddy and filled along the northern boundary as a broad dyke. This dyke was designed to serve several purposes: to separate the neighbor's paddy from ours and to site vegetable beds along the canal, as well as to provide a walking trail and an agroforestry strip along the boundary.

Paddy soil, such as we used to develop the dyke, when dug up by a backhoe and having its horizons mixed and exposed to the sun, turns into an abominable concrete-like surface, completely unsuitable for growing plants in the first year. Taro, maize and a range of other plants - even with the help of compost - died or were completely stunted. Mung beans survive in the moister cracks and do moderately well. Sunn hemp, when the rock-hard clods are broken up and shallow holes drilled, grow remarkably well. We planted the area with sunn hemp and a variety of beans - rice, black and mung - in the heart of the dry season and the strip greened up with the help of daily early watering, reduced to twice a week after the plants were established.

The paddy experience was a different matter. In January, after completing the earthwork discussed above, we hired a tractor to dry plow the paddy and erase the backhoe tracks. We briefly flooded and drained the paddy and then broadcast a variety of legumes in test strips, specifically sunn hemp as well as five types of bean (rice, mung, black, lablab, winged). Germination was not very good, and I was concerned that the surface of the soil was drying out and inhibiting proper germination. In retrospect, though, I was too hasty. In fact, the subsurface soil was probably still saturated.

Consequently, we flooded the paddy again and most of the beans failed to germinate. The black and mung beans sprouted but failed to grow beyond small plants. The soil was still likely too wet and the roots were affected.

On the other hand, on some ridges in the roughly leveled paddy the sunn hemp did well. This led me to believe that the problem was too much water, not too little, and a second broadcast of just sunn hemp resulted in about 40 percent cover.

At the same time, I set aside two test plots in a higher corner of the paddy, each about 5 m x 5 m (16.4 ft. x 16.4 ft). In one plot I thickly broadcast sorghum (for organic matter) and sunn hemp. In the other, we drilled mixed holes of the two crops. Both plots received no fertilization. Each was watered by hand daily for about seven days and then twice a week for about two more weeks.

AN 3 Crotalaria juncea, a promising green manure crop for the tropics 3In each plot the sunn hemp grew thickly - more thickly in the broadcast plot - with better sorghum establishment in the drilled plot. After solid establishment in about three weeks we stopped watering. This was in the driest period of the year. With the exception of a few unseasonably wet days in late February and March, there was no rain. 

Using a hand sickle, I cut back both the sunn hemp and sorghum to about .5 meters (19.7 in.) when the former began to flower in just under six weeks, setting aside about five plants in each plot for seed. I allowed the biomass to fall as mulch, and repeated this twice over the next six weeks. We used a mechanical weed "whacker" to cut back the sunn hemp in the rest of the paddy. I also did a one-time harvest of seed from the remnant plants.

While weed growth in the test plots was greater than in a dry adjacent field, the sunn hemp did help suppress weed growth, especially when compared with another paddy section that was fertilized with compost and planted with pumpkins.

In May we dry plowed both the paddy and the broad dyke after the latter was spread with some sandy soil and horse manure. In both cases, many of the mature sunn hemp pods that had previously been unharvested were incorporated into the soil. Eventually, in both sites, a lawn of young sunn hemp emerged thick and vigorous before we could plant any new seed, although in the paddy this was true only in the higher, drier corner where the test plots were situated.

As the plants began to flower, they were cut back along the dyke. Some were collected for compost, while some were allowed to fall as mulch. In the paddy, we flooded and plowed to prepare for the rainy season rice crop.

AN 3 Crotalaria juncea, a promising green manure crop for the tropics 4By mid-October 2009, around one month shy of harvest time, the rice fields were yielding approving comments by local rice farmers -- and this with no use of synthetic fertilizers for three years. And along the dyke, in what will become a 50 sq m strip of vegetable beds, we are sowing a long "mulch bed" of mung beans, sunn hemp and sorghum (for organic matter) amidst the re-sprouting sunn hemp (although in some areas the mulch is too thick for broadcasting).

The plan in the paddy is to broadcast sunn hemp after (or just before) the rice harvest in December. It will be cut back in late January or early February before growing a dry season crop of rice, soybeans and other things. For that to happen, we have set aside clumps of sunn hemp in the "mulch" beds for seed. The bright yellow flowers and pollinating insects that visit them are a welcome sight around the farm.

[Editor: Jeff can be contacted at tjeffrutherford@gmail.com.

For more information about sunn hemp, readers may refer to ECHO's Green Manures and Cover Crops page.

Additionally, sample seed packets of sunn hemp can be ordered (when available) from the Leguminous Ground Covers and Green Manures section of ECHO's seed bank page.


University of Hawaii:

Tropical Forages:

FAO: http://www.fao.org/ag/AGP/AGPC /doc/GBASE/data/pf000475.html
Purdue University: http://www.hort.purdue.edu/newcrop/proceedings1996/v3- 389.html

Treadwell D.D. and M. Alligood. 2008. Sunn hemp (Crotalaria juncea L.): A summer cover crop for Florida vegetable producers. Horticultural Sciences Department, Florida Cooperative Extension Service, Institute of Food and Agricultural Sciences, University of Florida. Accessed October 27, 2009. http://edis.ifas.ufl.edu/HS376