National Institute of Genetics, Mishima, 411 Japan
To look into the evolutionary dynamics of plant-pathogen systems in natural and agro-ecosystems, we have studied intrapopulational variations in the resistance to bacterial blight (BB) in the common wild rice and cultivated rice in farmers' fields. Six populations of Oryza rufipogon (Thailand and Bangladesh) and 12 populations of O. sativa (China, Bangladesh and Japan) were investigated for their reaction to four Japanese races of Xanthomonas campestris pv. oryzae (I, II, III, and IV). The plants examined in each population were derived either from a bulk sample or from lines which were randomly taken from natural populations. Their reactions to the four pathogen races were each classified as resistant (R) or susceptible (S), and a total of 15 different reaction types were observed.
Table 1 shows the frequency distribution of reaction types found between and within host populations. In Thai wild rice, two annual populations which are adapted to shallow water conditions showed a high degree of diversity in reaction types, while two perennial populations which are adapted to deepwater conditions consisted mainly of RRRR type. It should be noted that, of the two polymorphic annual populations, NE3 showed a hich frequency of SSSS type. This population was growing in a flat area which is completely parched in the dry season. The NE4 population contained many RRRR plants. This exceptional annual population was growing in a depression where the lower site retained water throughout the year. The two wild populations from Bangladesh, that were weedy types growing in deepwater rice field, were highly polymorphic containing many resistant individuals.
The farmers' landraces were generally polymorphic as compared with im-
Table 1. Intra-populational variation in resistance to four Japanese races
of Xanthomonas campestris pv. oryzae in wild rice and farmers' fields
===============================================================================
Species/Country/ Reaction type1 No.of Diversity
Population plants(p) index
or Variety 4R______3R________________2R_________1R____OR or
lines(l) H/2
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
===============================================================================
O.rufipogon
Thailand
NE3(annual) 3 3 5 2 1 1 12 1 22 50(p) 1.43
NE4(annual) 39 7 3 5 1 1 3 8 67(p) 1.34
CP20(perennial)33 1 1 1 36(p) 0.39
NE88(perennial)43 2 1 1 47(p) 0.36
Bangladesh
Jhora I(weedy) 4 1 1 3 1 1 11(p) 1.26
Jhora II(weedy)6 1 1 4 4 1 1 1 19(p) 1.77
O.sativa
China
Ch112(lowland)10 1 6 17(p) 0.85
Ch113(lowland) 24 6 30(p) 0.50
Ch114(upland) 2 4 4 10 20(p) 1.22
Bangladesh
Dhaki(aman) 7 3 4 1 15(p) 1.22
Bilpara(aman) 17 4 1 1 23(p) 0.78
Purbachi(boro) 7 2 1 10(l) 0.80
Iriatom(boro)3 10 10(l) 0
Pajam(boro)3 10 9 19(l) 0.69
IR8(boro)3 1 9 10(l) 0.33
BR3(boro)3 10 10(l) 0
BR14(boro)3 10 10(l) 0
Japan
Tanegashima 4 28 32(p) 0.38
===============================================================================
1) Reaction to Japanese pathogen races I, II, III and IV
1: RRRR 2: RRSR 3: RRRS 4: SRRR 5: RSRR 6: SRRS 7: RRSS 8: RSSR
9: SRSR 10: RSRS I 1: SSRR 12: RSSS 13: SRSS 14: SSSR 15: SSSS
2) H'= - sum(p\i\) In p\i\, pi stands for the frequency of plant number
showing i-th reaction type. 3) Local improved or high-yielding varieties,
others are all landraces.
proved varieties (5 populations from Bangladesh) which were highly or
moderately resistant. Chinese populations collected in a hilly area in Yunnan
were relatively susceptible. On the other hand, two aman landraces from a
deepwater area in Bangladesh were resistant. A boro landrace (Purbachi) grown
in the dry season in the same deepwater area tended to be slightly
susceptible. One Japanese old variety which has been traditionally
cultivated in Tanegashima was highly susceptible to Japanese races.
It was demonstrated that, in wild rice and landrace populations, grown in deepwater tend to have a higher frequency of resistant genotypes than those grown in drier conditions. Deepwater conditions which favor epidemics of bacterial blight must have operated as a selection pressure. Loss of resistance in host populations which were moved to drier or disease-free environments may be explained by assuming "fitness cost" in the host plants. As to the degree of intra-populational variation, diversity in BB resistance is not correlated with isozyme diversity which is mainly determined by mating and propagating systems (Morishima and Barbier 1990).
A number of factors affect resistance structure of host populations as reviewed by Burdon (1987). It is difficult to explain our results as the variation pattern observed represents the interaction between host plants and Japanese races of pathogen, not the races prevailing in the natural habitat of host plants. Yet, the above results seem to reflect the consequences of certain kinds of natural selection. The genetic basis for BB resistance operating in our study populations cannot be elucidated well. A preliminary study, however, has suggested that at least one dominant gene which confers general resistance to all four Japanese races segregates in Thai wild populations. Probably because of its wide spectrum of resistance, BB resistance genotypes detected in the present study must have been subjected to selection in their natural habitats.
References
Burdon, J.J., 1987. Diseases and Plant Population Biology. Cambridge University Press, Cambridge. pp. 208.
Morishima, H. and P. Barbier, 1990. Mating system and genetic structure of natural populations in wild rice Oryza rufipogon. Plant Species Biol. 5: 31-39.