SRI farmer Alkalifa Ag Insaye in the village of Horogoungou found a rice plant with 161 tillers and 160 panicles (see picture) in his SRI plot. To his surprise, it was a plant from an indigenous variety that had ‘contaminated’ the seeds of the improved variety. Only improved varieties are grown in irrigated agriculture. Indigenous varieties are cultivated in recession agricultural systems or as rainfed crops in landscape depressions, where rainwater collects. Yields are low with a regional average of only 1500 kg/ha, and sometimes rice crops in the depressions fail if rainfall is not sufficient.
 
 
October 31, 2008
Record Tillering with Indigenous Rice Varieties
 
 
 
 
In Horogoungou village, the indigenous variety was slightly taller and matured more quickly than the tested variety BG-90-2 and thus could easily be spotted in the field (see arrows in picture below). Both farmers and field agents noticed that the indigenous plants produced higher tiller numbers in the SRI plots than in the control plots.
 
To test this, we harvested a total of 16 randomly selected indigenous variety plants from four SRI plots and 16 indigenous variety plants from four control plots in the village of Horogoungou. We also harvested 9 plants (one sq. meter) of the indigenous variety, planted under traditional rainfed conditions (see pictures further below).
a) the 161 tiller plant in the field (observe how widely the tillers are spread: framed by red lines), b) harvested, and c) shown as “trophy” by happy and proud farmers and Africare field agents.
Alkalifa harvests individual indigenous plants in his SRI field. Red arrows indicate indigenous variety plants growing among the improved variety. Africare field agent Ibrahima holds up two indigenous rice plants harvested in the SRI plot.
We counted the tillers on each of the harvested plants, and threshed the 16 plants from the SRI and the control plots immediately after harvest. Finally, we weighed the harvested grain and measured its moisture content.
Our impressions of higher tillering under SRI were confirmed. Nine of the 16 plants exceeded 80 tillers and panicles per plant. Average tiller numbers were 79 for SRI, 46 for the control, and 22 for the rainfed rice. Tiller numbers for each of the harvested plants are shown in the scatter plot below. The percentage of fertile tillers (tillers with panicles as compared to total number of tillers) was 98.6% for SRI, 96.6% for the control and 100% for the rainfed rice.
 
Indigenous varieties planted under the traditional rainfed cropping system in shallow landscape depressions. A farmer holds up two harvested plants (see red circle and arrow). Notice the difference in tillering with the SRI plants shown above by Ibrahima.
Tiller numbers of 16 randomly harvested plants of a local rice variety in SRI and control fields, and from 9 plants under rainfed conditions.
 
 
We measured the grain yield and calculated it for nine plants: rainfed rice produced 209g, the control plot rice 521 g, and the SRI rice 1004 g. If we assume a density of 9 plants/m2, this indicates a yield of 2t/ha, 5.2 t/ha and 10t/ha for the three rice-cropping systems, respectively. We are aware that these are assumptions: although plant density was higher in the test plots (16 plants/m2), the single indigenous plants were neighbors to less profusely tillering improved variety plants, thus may have had more space to develop the high tiller numbers. Nevertheless, the results provide us with some important indications and call for a more thorough test next year.
 
The Mali Government agricultural service does not recommend using indigenous varieties under irrigation based on two characteristics: 1) susceptibility to lodging (plants fall over easily), and 2) easy shattering (grains fall from the panicle). Nevertheless, Africare field agents and farmers observed that plants harvested under irrigated conditions did not seem to lose their grains easily.
 
Farmers also confirm that traditional varieties grown under irrigated conditions show higher yields  compared to the same varieties grown under rainfed systems. They are not grown under irrigation because the market price of these varieties is lower: 200 CFA/kg, compared to 300-350 CFA/kg for the improved variety. On the other hand, farmers say the local varieties taste better by far!
 
Farmers are extremely interested in testing the indigenous varieties using SRI techniques in the irrigation perimeters next year, and suggest testing deepwater rice varieties as well. They know of 4 or 5 rainfed and a similar number of deepwater rice varieties. We decided to collect seeds of the different varieties in order to test them next year.
 
Growing these indigenous varieties of rice is highly important to maintain biodiversity. Our project zone is part of the Middle Niger Delta, which is known to be among the earliest sites for domestication of African rice (Oryza glaberrima). Favored for its resistance to drought, disease, and pests, its nutty flavors, and filling properties, the rice has spread from the Niger Delta over broad regions of western Africa during the past 1000 years and more. See Searching for the origins of African rice domestication by Shawn Sabrina Murray  
 
As recommended by the farmers, we plan to include deepwater rice varieties in the SRI tests under irrigated conditions next year. These pictures taken from a pirogue show 1) an almost mature deepwater rice field, 2) a field where pockets from transplanting are still easily visible, and 3) a panicle from an indigenous black deepwater rice variety.