3. Semidwarfing genes in rice germplasm collection

A single recessive gene (sd\1\) from 3 independent Chinese sources-Dee-geo-woo-gen, Ai-jiao-nan-te and Ai-zai-zhan, has led to the development of semidwarf rice varieties in Taiwan and mainland China, several Asian Countries, and at IRRI. This gene has lent great impetus to rice production increases in many Asian countries.

Ever since IRRI began its research operations in 1962, we have been interested in the genetic control of the improved plant type and the sources of semidwarfing genes. This article summarizes our decades of search for other sources of semidwarfs in the rice germplasm collection of IRRI.

We have identified about 145 short-statured accessions from the IRRI germplasm collection and have made more than 200 crosses to date. Each of them was crossed with Taichung (Native)l, IR8, IR20 or IR36 to test the allelic relationship of their semidwarfing gene(s) with that of sd\1\.

From a concurrent planting and comparison of the two parents, F\1\ plants and F\2\ progenies, the gene systems in the new semidwarfs were classified into 3 categories:

1. Identical with sd\1\ locus: F\1\ plants and all F\2\ plants are semidwarfs, although the effect of modifying genes may be detected.

a) Purbachi; derivatives of Ai-zai-zhan and Al-jiao-nan-te (China). b) Induced Mutants of I-kung-bau and Keh-tze (Taiwan). c) C53-39, Khunnaywayin(S) and (P), Khunni Shay, Nga-Kywe (Burma). d) Culture 147, Culture 155, T141/Baok 360, ARC 5929 and other ARC semidwarfs, Crm 13-3241, BM 13, CRHP8, Jikkoku/Shiranui (52-37; 52-102), Shiranui/Seraup Kechili 55-296, Shiranui/Gionchiew 61-8, CNM 25, C12329, IET 2895, New Sabarmati, Barmda 21, 23 and 828, KH 863 (India). e) "California 2" (Philippines). f) Calrose 76, M9, M101, M302, S201 and LA 110 (U.S.A.).

II. Sharing a compound locus with sd\1\: F\1\ plants are semidwarf; great majority of the F\2\ plants are semidwarfs; only a small number of F\2\ plants (less than 1%) are intermediate-tall or tall. a) Fukunishiki, Kochihibiki and Reimei (Japan). b) Gora (NCS 16), Synthetic Sativa and TR17 (India). c) Sekarmandi (Indonesia).

III. Non-allelic gene(s): F, plants are taller than either parent: distinct segregation for plant height is observed in the F\2\ populations.

a) Cl 9649/Cl 9722, Cl 9858, CP231/SLO-17 (B5580AI-15), Double Dwarf 1, Intermediate Dwarf, Short Straw Dawn, Short Straw Starbonnet (U.S.A.). b) CN 242d\3\, Tainan-5-mutants 72-534, 72-536; Tainan-5-mutants 73-66, 73-75, 73-111, TNA 761-1-148-157, Tainan 6 mutant 772039, 772040 (Taiwan). c) 2243-85F and IRAM 2165 (Ivory Coast). d) Culture 854, Culture 956, d\6\, d\7\, d\8\, d\10\, Kalimonch Dwarf Mutant, Mutant 65, P-3 dwarf (India). e) Nadula Dwarf (Fiji). 0 Fanny Dwarf (France). g) GS 1649A (China via Thailand). h) K8 mutant (Sri Lanka). i) Hoyoku Uapan).

The above tabulation, however, does not include many semidwarfs whose pedigrees may be traced to one of the known Chinese semidwarf sources (sd\1\).

Plant height, being a quantitative trait, is subject to environmental influence of temperature, photoperiod and soil fertility. Many short-statured varieties and lines from temperate zones tend to be shorter and earlier at IRRI than in their home habitat. Hence, the classification of the progenies may be affected by genotype x environment interactions.

Genetic incompatibility between ecogeographic races may also lead to aberrant segregation. Cytoplasmic-nuclear interaction may be detected in some reciprocal crosses.

Additional genes inhibiting tall height may affect the expression of F\2\ plants which were reported to have the sdi gene.

Some of the accessions may have been outcrossed or mislabelled when handled by workers in a different country. Some "dwarfs" are more likely semidwarfs.

The sd\1\ gene appears to belong to a readily mutable locus so that many spontaneous and induced mutations had occurred at the same site. Its compound nature needs to be further studied by expanded F\2\ populations, say, over 10,000 F\2\ plants per cross.

Among those semidwarfs who have gene(s) non-allelic to sd\1\, the great majority of them were extremely poor in agronomic characters, i.e., weak growth vigor, shorter height than is desired, poor tillering, unattractive panicles, poor panicle exsertion, poor grain filling, hard threshability, and miniaturized grains. Only the following strains appeared relatively better: CP231/SLO-17 (Acc. 6993), Culture 854, Culture 956, D66, P-3 dwarf, Tainan 5 mutants (72-534, 72-536), Reimei, and Hoyoku. But none of them can compare with IR8, TN1 or IR36 in growth vigor and grain yield under Los Banos conditions.