proper research could perhaps triple production of deep-water rice in the Niger's inland delta. This is one area of research that can do something toward reducing hunger in one of the regions of Africa most in need of help.
Although the current African types shed grain more readily than the Asian ones, some improvements have been bred into dryland varieties. Additional research emphasizing seed shattering could make a big difference. Because the gene for nonshattering is recessive, the selection of nonshattering types should be rapid, and true breeding should be immediate. Other improvements might include selection for resistance to disease. This resistance exists in the various genotypes, and the major problem is not to lose these local types as Asian rice spreads even further. For the uplands, any form of rice must resist blast and sheath blight. All types must also resist rice yellow-mottle virus; some local cultivars already do.
For areas dependent on seasonal flooding, varieties must resist lodging and respond to fertilizer; the transplant types must tolerate widely varying periods of growth in the nursery (while farmers await the onset of the unpredictable natural flooding).
Researchers are at present "mapping" the chromosomes of both African and Asian rice, identifying the portions that control various features of the plant.13 This powerful modern technique will "jumpstart" the genetic improvement of African rice (see box, page 34). Perhaps it could also facilitate the transfer of useful genetic material between the two.
Oryza glaberrima Steudel
Oryza barthii ssp. glaberrima
English: African rice, glaberrima rice
French: riz pluvial africain, vieux riz, riz africain, riz flottant
Cameroon: erisi (Banyong)
Guinea: Baga-malé, malé, riz des Baga
Mali: Issa-mo (river rice), mou-bér (great rice)
Sierra Leone: kebelei, mba, mbei (Mende), mala (Kissi), Kono, pa (Temne)