Table Of ContentsNext Page

Development of Lowland Rice from the Interspecific Cross of Oryza Sativa and O. Glaberrima

Koichi Futakuchi1, Monty P. Jones2 and Olu Osiname3

1 WARDA―The Africa Rice Center, c/o ICRISAT, BP 320 Bamako, Mali, E-mail
Forum for Agricultural Research in Africa, PMB CT 173 Cantonments, Accra, Ghana, E-mail
WARDA―The Africa Rice Center, Nigeria Station, c/o IITA, PMB 5320 Oyo Road, Ibadan, Nigeria, E-mail


After the success of developing interspecific Oryza sativa x O. glaberrima progenies for upland, WARDA has extended the target of the interspecific breeding to lowland. The interspecific crosses between the two species were initiated in 1996. After two backcrosses to the O. sativa parents, pedigree selection in lowland conditions started in 1998, of which major targeted characteristics were vegetative growth of O. glaberrima and reproductive growth of O. sativa associated with weed competitiveness and yielding ability, respectively. Expected interspecific progenies have been identified during the selection. Some interspecific progenies showed better growth (height and dry weight) and more allocation of dry matter to the leaves than the O. sativa check variety at the seedling stage, as did the O. glaberrima check. All interspecific progenies tested had the same or higher levels of yield than the O. sativa check varieties in both irrigated and rainfed lowlands. Yield of two interspecific progenies was more than double compared to that of the best O. sativa check in irrigated lowland.

Media summary

Lowland rice has been developed from the wide cross of Oryza sativa and O. glaberrima. Some of them depicted promise in weed competitiveness and yield.

Key Words

Oryza species, SPAD, West and Central Africa, Wide cross.


Oryza glaberrima Steud. is the other cultivated Oryza species than O. sativa L. and has been cultivated in West and Central Africa (WCA) for more than 3500 years (Jones et al. 1997b; Singh et al. 1997). Since the species has unfavourable traits in yield formation, i.e. few secondary branches in a panicle, grain shuttering and poor resistance to lodging, its growing area has been decreasing. However, many O. glaberrima lines are known to have resistance to various biotic and abiotic stresses in WCA (Jones et al. 1997b; Johnson et al. 1998; Jones and Singh 1999; Futakuchi et al. 2001) and are important as genetic resources to develop suitable rice varieties for resource poor farmers in WCA, who are suffering from low yielding due to multiple stresses in their fields. WARDA initiated an interspecific hybridisation program in 1992 to combine characteristics of O. sativa associated with high yielding ability and resistance of O. glaberrima to the major constraints in the sub-region and succeeded in developing fertile interspecific progenies in 1994 (Jones et al. 1997a). The target of the program was upland farmers because major ecology for rice cultivation is upland which accounts for about 60% of the total rice cultivation area in WCA (Jones 1999).

Rice cultivation in more robust lowland areas has considerable potential for further development; the higher yield potential of these lowland areas compared to the uplands could play a key role in assisting countries in WCA to achieve rice self-sufficiency. Since O. glaberrima also showed promise for use in lowland breeding programs, due to their strong resistance to important lowland constraints such as weeds, submergence, iron toxicity, rice yellow mottle virus and African gall midge, WARDA has commenced the selection of interspecific progenies for lowlands. This paper reports preliminary observation of the progenies in relation to yield and other traits targeted in the selection.


Interspecific hybridisation and selection

The main objective of this selection was to develop rice cultivars which possess O. glaberrima type vegetative growth (weed-competitive) and O. sativa type reproductive growth (characteristics associated with high yields). Therefore, the criteria of the “O. glaberrima type” are vigorous growth at the seedling stage and high tillering, thin, droopy and light green leaves, and ground coverage at the early growth stage (Johnson et al. 1998; Dingkuhn et al. 1999), whereas those of the “O. sativa type” are resistance to lodging and grain shuttering, and a similar panicle type with O. sativa.

The wide cross was initiated in 1996. The F1 progenies were backcrossed to the O. sativa parents. After two backcrosses, 35 BC2F1 progenies were obtained from 22 crosses. In 1998, 11 BC2F1 progenies derived from 7 crosses were selected due to better fertility (the number of fertile grains produced) and subjected to pedigree selection in lowland conditions. Simultaneously, several promising progenies, in terms of target characteristics, were sampled after F5 generation and passed through an anther culture process (Jones et al. 1997a) in order to accelerate their fixation.

(a) Seedling vigor

On the WARDA experimental fields at M’b near Bouak, Cte d’Ivoire, 12 interspecific progenies (WAB1159 and WAB 1031 series) were tested in the 2002 wet season with 2 checks: Bouake 189 and CG 14. WAB1159 series were developed from the cross of WAB56-50 (O. sativa, a japonica type) and Shawhon (O. glaberrima). WAB1031 series were derived from the cross of WITA 7 (O. sativa, an indica type) and CG 20 (O. glaberrima). Bouake 189 is a popular O. sativa lowland variety in Cte d’Ivoire and CG 14 is an O. glaberrima line. Seeds were sown on a semi-irrigated nursery without fertilizer. Seedlings were sampled on 22 days after seeding (DAS), and dry weight, plant height and SPAD values on the upper most fully expanded leaf were determined.

(b) Yield in the humid zone of WCA

Twelve interspecific progenies (WAB1159 series) were evaluated on the experimental fields of WARDA Nigeria Station in Ibadan, Nigeria in the 2002 wet season with 4 checks: ITA 230 and WITA 4 (officially released in Nigeria), and TOX 4004-43-1-2-1 and TOX 4303-13-3-1-1-2 (elite lines in Nigeria). The materials were grown in both irrigated and rainfed lowlands in a randomised complete block design with 3 replications. Seedlings were raised in a nursery and transplanted, 21 DAS, in a square-planting pattern with 20 cm between hills and 2 or 3 seedlings per hill. The N-P-K fertilizer was applied at the rate of 60-30-30 kg/ha. The fields were manually weeded two times after transplanting.


Pedigree selection

WAB1159-2-12-11 (F5) had a highly bushy canopy, comparable to that of O. glaberrima. It had wide, thin, light green, droopy leaves and high tillering ability. Since segregation was still observed at the F5 generation, various types of plants could be developed from its progenies in terms of: (i) growth duration - from 119 days (WAB1159-2-12-11-6-6) to 149 days (WAB1159-2-12-11-2-2); (ii) plant height - from 110 cm (WAB1159-2-12-11-2-3) to 150 cm (WAB1159-2-12-11-6-7); and (iii) relative exertion of panicles from the flag leaf - various. All progenies of WAB1159-2-12-11 retained its O. glaberrima type vegetative growth and O. sativa type panicles in the later generations. In 2002, most populations at the F8 generation were homogenous but fixation should be confirmed in the next generation. In addition to the progenies of WAB1159-2-12-11, 5 fixed interspecific progenies, e.g. WAB1159-4-10-15-1-5, were obtained in 2001. Since they have large panicles, they had fewer tillers than the progenies of WAB1159-2-12-11. Their yields in the wet season ranged from 5.5 to 8.5 t/ha with the 100 kg/ha of nitrogen input.

Seedling vigor

The seedling vigour (plant height and total dry weight) of CG 14 (O. glaberrima) was much higher than that of Bouake 189 (O. sativa) (Table 1). CG 14 also had a low SPAD value and a high ratio of leaf dry weight to total dry weight compared to Bouake 189 (Table 1). A lower SPAD value corresponds to a thinner and lighter green leaf, and a higher ratio of leaf dry weight to total dry weight means that dry matter is more allocated to the leaves, which are photosynthetic organs. These characteristics could be associated with rapid leaf expansion, and consequently a high growth rate in the next stage. CG 14 will therefore have better weed competitiveness than Bouake 189. Based on all four traits shown in Table 1, WAB1159-2-12-11-6-8 and WAB1159-2-12-11-6-10 are potentially the best weed competitors among the interspecific progenies. Other than the SPAD value, these two lines had more favourable values than Bouake 189. Furthermore, their dry weight is much heavier than even CG 14. The SPAD value of WAB1159-1-12-11-2-10 was as low as those of CG 14 and would be expected to have a high growth rate at the next stage.

Yield in the humid zone in WCA

Figure 1 shows the yields of interspecific progenies in both irrigated and rainfed lowlands in Ibadan in the 2002 wet season. Three inperspecific progenies gave significantly higher yield than WITA 4, which had the highest yield among the checks. Two of them, WAB1159-2-12-11-6-7 (V1) and WAB1159-2-12-11-2-10 (V2), gave more than double the yield of WITA 4. Their yielding ability could be quite high compared to the standard O. sativa varieties. In the rainfed lowland, varietal differences in yield were not so clear. However, WAB1159-2-12-11-2-10 (V2) yielded significantly higher than TOX 4303-13-3-1-1-2 and ITA 230. In both irrigated and rainfed lowlands, no interspecific progeny had significantly lower yield than any of the checks. All the interspecific progenies had acceptable yield levels.

Table 1. Growth characteristics of seedlings in a crowded nursery.


Plant height




Inperspecific progeny
































































CG 14





Bouake 189






Level of significance





LSD (P=0.05)





1. TDW=Total dry weight; 2. LDW=Leaf dry weight.

**, P<0.01

Figure 1. Yield of interspecific progenies in irrigated and rainfed lowlands in Ibadan, Nigeria. V1-V12 = interspecific progenies; C1 = TOX 4004-43-1-2-1; C2 = TOX 4303-13-3-1-1-2; C3 = ITA 230; C4 = WITA 4.


Selection of interspecific O. sativa x O. glaberrima progenies has been conducted at WARDA in order to develop lowland varieties possessing both weed competitiveness of O. glaberrima and characteristics of O. sativa associated with high yielding. Some of the interspesific progenies could be better weed competitors than the popular O. sativa variety judging from the characteristics at the seedling stage. Yield of all interspecific progenies tested were in acceptable levels in comparison with the check varieties in both irrigated and rainfed lowlands.


Dingkuhn M, Johnson DE, Sow A and Audebert AY (1999). Relationships between upland rice canopy characteristics and weed competitiveness. Field Crops Research 61, 79-95.

Futakuchi K, Jones MP and Ishii R (2001). Physiological and morphological mechanisms of submergence resistance in African rice (Oryza glaberrima Steud.). Japanese Journal of Tropical Agriculture 45, 8-14.

Johnson DE, Dingkuhn M, Jones MP and Mahamane MC (1998). The influence of rice plant type on the effect of weed competition on Oryza sativa and Oryza glaberrima. Weed Research 38, 207-216.

Jones MP (1999). In ‘World Food Security and Crop Production Technologies for Tomorrow’. (Ed. T. Horie, S. Geng, T. Amano, T. Inamura and T. Shiraiwa) pp. 57-64. (Kyoto University, Kyoto).

Jones MP, Dingkuhn M, Aluko GK and Semon M (1997a). Interspecific Oryza sativa L. x O. glaberrima Steud. progenies in upland rice improvement. Euphytica 92, 237-246.

Jones MP, Mande S and Aluko K (1997b). Diversity and potential of Oryza glaberrima Steud. in upland rice breeding. Breeding Science 47, 395-398.

Jones MP and Singh BN (1999). In ‘World Food Security and Crop Production Technologies for Tomorrow’. (Ed. T. Horie, S. Geng, T. Amano, T. Inamura and T. Shiraiwa) pp. 133-136. (Kyoto University, Kyoto).

Singh BN, Fagade S, Ukwungwu MN, Williams C, Jagtap SS, Oladimeji O, Efisue A and Okhidievbie O (1997). Rice growing environments and biophysical constraints in different agroecological zones of Nigeria. Meteorological Journal 2, 35-44.

Top Of PageNext Page