5. Genetic erosion in wild and cultivated rice species
        H. Morishima and H. I. Oka
        National Institute of Genetics. Mishima. 411 Japan
        Genetic diversity as observed in landraces of rice and its wild relatives enables the plants to evolve and differentiate into various cultivars to adapt to different environments. Recently, however, the diversity of gene pool has become depleted, mainly on account of extension of modern high-yielding varieties. Natural habitats of wild rice are also threatened by various development projects of the farmland.

    In many rice growing countries, the landraces carrying a vast amount of genetic diversity were distributed in remote villages. The number of landraces began to decline in 1970's when high-yielding varieties were introduced. Most of the old landraces are now available in certain gene banks only. not in the hands of farmers.

    For instance, in Japan, collection of native rice varieties was carried out by Prof. T. Nagamatsu of Kyushu University and some researchers of Ministry of Agriculture. Forestry and Fishery over the whole country, starting in 1962. and more than 1.000 varieties were collected. They are now preserved by the Institute of Genetic Resources. Kyushu University. Fukuoka. but not available in their original habitats. In Japan, there is no wild rice and the rice germplasm is relatively narrow because most of them are Japonica types introduced from China either directly or through the Ryukyu islands. Yet. the landraces carried considerable genetic variation. In Taiwan, some 100 landraces grown by the mountain tribes in different villages were collected by the Taichung Agric. Exp. Station in 1943. which were largely tropical Japonica types. Now, these are not available in the mountain villages.

    In China. Indochina and India where rice cultivars are more diverse, the landraces carry a tremendous amount of variations. For instance, rice varieties in Bangladesh are divided into three major seasonal groups: Aus. Aman and Boro. These are grown by the farmers in accordance with the water regimes for particular farming system, including mixed planting. The deepwater types like Rayada and Ashina were found to carry rare isozyme alleles (Glaszmann 1987). In recent years, these traditional varieties have been replaced by modem improved varieties associated with new cultural practices.

    Similarly, the number of local varieties has decreased in the Mekong delta. Vietnam. during the last two decades. Shift to monoculture has occurred also in Thailand. According to the observation of rice varieties grown in a village near Khon-Kaen by Miyagawa (1994). the extension of an improved glutinous variety RD6 rapidly replaced other traditional varieties, as shown in Table 1.
 

    Populations of the common wild rice (Oryza rufipogon) are also declining in many places. There were three small populations of wild rice in village Pate. Taoyuan. Taiwan. each of which was in natural streams about 1 km apart from one another. They became extinct around 1975 as reported by Kiang et al. (1979). The plants which naturally hybridized with cultivars could not endure competition with Leersia hexandra which invaded into the habitat after some change in the water regime. Since 1980. H. I. Oka attempted to restore the wild-rice plants in one of the original sites using the original seeds, but his trial was unsuccessful. In 1986. a strain not contaminated much by hybridization (W1623) was planted, which appeared to be more successful in the 5-year period until 1991. This suggests that the hybridization with cultivars is the major cause of deterioration of Taiwan wild rice.

    Since 1983. the senior author and coworkers have selected seven fixed sites in the northern suburbs of Bangkok for observation of the common wild rice and visited the sites regularly. Of the seven populations, four were annual, two were perennial and one was a weedy type. An annual population NE2 which was monitored since 1975 declined and became finally extinct around 1987 on account of increasing growth of perennial weeds possibly because of increased water supply to the land in the dry season. Other three annual populations, after fluctuation in population size. became extinct around 1990 on account of habitat destruction caused by road-widening. A perennial (NE88) and a weedy (CP20) population also became extinct for the same reason. Another perennial population (CP24) has survived although has declined possibly because of water pollution.

    In the northern suburbs of Bangkok, the nearest stands of the common wild rice were at Bangkhen (about 10 km from city center, near the Rice Research Institute) observed in 1957. near the entrance of International Airport (about 20 km) observed in 1970. near Klong-Luang Rice Exp. Station (about 35 km) in 1975. and then there was the CP20 site near Ayuthaya (about 70 km. mentioned above). It remains unknown how urbanization damages wild plants like the wild rice.

    In addition to germplasm preservation in laboratories (ex situ conservation), in situ conservation of wild populations is desired. It needs administrative solution of rural social problems and budgetary support, so as to keep an area with wild rice safe from human encouragement. We do not have enough experience in this work (cf. Vaughan and Chang 1992). In general, the perennial populations of wild rice are more variable within each than
 
 

Table 1. Decrease of traditional rice varieties observed in a village near Khon-Kaen, Thailand (after Miyagawa 1994)

Year No. of varieties
planted
%area of RD6
to total
1981 27 0.2
1983 24 0.3
1991 4 75.4

    annual populations, while the annuals are more variable from each other. This trend was recognized in both isozymes and phenotypic characters (Oka 1988. p.70-72). This was also found in the comparison of annual, perennial and weedy populations of the present study-sites (Table 2). The perennial wild-rice outcrosses more (30-50%). while the annual wild rice has relatively lower outcrossing rate (10-20%). This difference in the breeding system results in different population structures.

    However, the within-populational diversity in the reaction to races of bacterial blight was greater for the annual than for the perennial populations (Table 2. shown by H'=-Sum pilnpi, Morishima 1994). The reason for this is not clear.

Table 2. Comparison of intra-populational variability between perennial and
        annual types of wild rice observed
 

Population Gene diversity
H=l/nSum (1-Sum xij2) Mean for 9 isozyme
loci
Coef. of
variation SD/Mean
Mean for 6 characters
Diversity index
H'=-Sum pilnpi Reaction to
bacterial races
Annual
NE3 0.208 0.211 1.43
NE4 0.147 0.214 1.34
Perennial
NE88 0.350 0.308 0.36
Weedy
CP20 0.327 0.263 0.39

    When the habitat is disturbed, some individuals of populations may move to and colonize new niches by seeds or by stublle pieces. Annual populations have higher possibility of migration than perennial ones. Even in seemingly stable populations, extinction and colonization at subpopulation level takes place. According to Slatkin (1977). the probability of extinction of population fragments (e0) is expected as: X=(l-e0) T where X is the proportion of extant or still present quadrats and T is the duration of observation in number of years. Adopting this formula, the probabilities of extinction of respective quadrats under observation were estimated and averaged tor each population are shown in Table 3. The annual populations have higher rate of local extinction than perennial ones.

Table 3. Propagating system and local extinction rate of wild-rice populations
            estimated at Fixed study-sites in Central Plain. Thailand
 

Population No. of quadrats studied Proportion of seedlings to total plantsa Ecotype Local extinction rate. eob
NE3 6 100±0 Annual .040
NE4 7 99.6 ± 0.5 Annual .055
CP20-1 3 72.0±2.3 Weedy .078
CP30-2 4 19.8 ± 8.7 Weedy .000
NE88 4 6.3 ± 8.2 Perennial .004

a) Including seedlings and vegetative propagules in early rainy
        season, mean for 2-9 censuses.
b) X=(l-eo)T ,T=No. of years observed, X=proportion of remaining (extant) quadrats after T years
    (Slatkin 1977).

References

Glaszmann, J. C., 1987. Isozymes and classification of Asian rice varieties. Theor. Appl. Genet. 74: 21-30.
Kiang, Y. T., J. Antonovics and L. Wu. 1977. The extinction of wild rice. 0ryza perennis formosana in
        Taiwan. J. Asia Ecol. 1: 1-9.
Miyagawa, S., 1994. Extension of improved variety RD6 and its impact on rice cultivation in rain-fed paddy
        area in Northeast Thailand. Korat Research 6: 148-160 (in Japanese).
Morishima. H.. 1994. Polymprphism of bacterial blight resistance in the populations of wild and cultivated rice:
        A lesson from natural ecosystems. In Toward enhanced and sustainable agricultural productivity in
        the 2000's. Breeding researches and biotechnology. Taichung Agric. Improvement Station. Tatsuen.
        Changhua. Taiwan. ROC. pp. 139-144.
Oka, H. I. 1988. Origin of Cultivated Rice. Elsevier/JSSP. Amsterdam/Tokyo.
Slatkin, M., 1977. Gene flow and genetic drift in a species subject of frequent local extinctions. Theor. Popul.
        Biol. 12:253-263.
Vaughan, D. A. and T. T. Chang. 1992. In situ conservation of rice genetic resources. Economic Botany. 46(4):
        368-383.