14. Cytogenetic relationships among the three species of Oryza latifolia complex

Three tetraploid wild species of Oryza, e.g., O.latifolia Desv., O. alta Swallen and O. grandiglumis Prod. have been described from Latin America. All three of them have CCDD genomes. Chatterjee (1948) and Tateoka (1964) considered them to be independent species. Oka (1961) surveyed the habitats of the three species. On the basis of his observations of many populations he considered the three taxa to be con-specific. Similarly, Sampath (1962), Kihara (1964) and Gopalakrishnan and Sampath (1967) suggested that the three species should be merged together.

In order to re-examine this problem we crossed O. latifolia from Mexico (Acc. No. 100914), O. alta from Surinam (Acc. No. 100888) and O. grandiglumis from Brazil (Acc. No. 100161) in all possible combinations without reciprocals. The hybrids were easily obtained and the crossability between species ranged from 25 to 59.4% (Table 1). Thus no crossability barriers amongst the three species were encountered. The hybrids of O. latifolia with other two species were completely pollen and seed sterile. The hybrids of O. grandiglumis and O. alta had 2.1% stainable pollen but were completely seed sterile (Table 1).

The modal chromosomal association in all the hybrids was 22II and 1IV. A small proportion of cells with 24II or 21II+1VI were also observed (Table 1). Gopalakrishnan and Sampath (1967) observed 24II at diakinesis in all hybrids and reported high spikelet fertility in the hybrids of O. grandiglumis and O. alta. The differences in their and our observations in respect to spikelet fertility of this hybrid are likely due to differences in the strains of two species used. The occurrence of a quadrivalent in hybrids may be due to the fixation of a reciprocal translocation in the strain of one of the species used in this study.

The three species are morphologically very similar. All are very vigorous, tall with sturdy stems, big panicles and large grains. They show gigas characteristics typical of tetraploids. They have no crossability barriers and their genomes are completely homologous. On the basis of F\1\ sterility these species might be considered independent but F\1\ sterility in Oryza is not a valid criterion for separating species. As is well known, many intra-specific crosses of O. sativa show high sterility. Similarly complete sterility was observed in the F\1\ hybrids of different ecotypes of O. officinalis complex although on the basis of morpological similarities, crossability and chromosome homology these taxa belong to same species comples (Jena and Khush 1985). Our data thus suggest that the three Latin American tetraploid species belong to one species, O. latifolia comlex.

Table 1. Crossability, fertility and chromosome pairing in hybrids of O. latifolia complex

Chatterjee, D., 1948. A modified key and enumeration of the species of Oryza Linn. Indian J. Agr. Sci. 18: 185-192.

Oka, H.I., 1961. Report of trip for investigation of rice in Latin American countries. pp. 1-40. National Institute of Genetics, Japan.

Sampath, S., 1962. The genus Oryza: Its taxonomy and species relationships. Oryza 1(1): 1-29.

Kihara, H., 1964. Need for standardization of genetic symbols and nomenclatur in rice. In IRRI (ed.), Rice Genetics and Cytogenetics, pp. 3-11, Elsevier, Amsterdam, London, New York.

Tateoka, T., 1964. Taxonomic studies of the genus Oryza. In IRRI (ed.), Rice Genetics and Cytogenetics, pp. 15-21, Elsevier, Amsterdam, London, New York.

Gopalakrishnan, R. and S. Sampath, 1967. Taxonomic status and origin of American tetraploid species of the series Latifoliae Tateoka in the genus Oryza. Indian J. Agr. Sci. 37: 465-475.

Jena, K.K. and G.S. Khush, 1985. Genome analysis of Oryza officinalis complex. RGN 2: 44-45.