吉田総合案内へ戻る．
本頁は“PPS 3(4)：375-376,2000”に掲載したものです．図表一部省略．

元鐘建・吉田智彦

Varietal Differences in Seedling Traits under the Low Dissolved 
Oxygen Level in Water-seeded Rice

Jong Gun Won and Tomohiko Yoshida
(Faculty of Agriculture, Kyushu University, Fukuoka 812-8581, Japan)

Key words : Low dissolved oxygen, Rice, Seedling, Varietal difference, Water-seeding.
Corresponding author: J.G. Won (won@agr.kyushu-u.ac.jp, Fax +81-92-642-2819, 
Kyongbuk Provincial ATA, 200 Dongho-dong, Buk-gu, Taegu 702-320, Korea).

      Rice seedlings from submerged seeds elongate rapidly until they reach the surface 
of the water. The leaves are long, narrow and delicate, but the roots are poorly developed. 
Therefore, the seedlings are not firmly anchored into the soil (Jones, 1933). The failure 
of the root development has been attributed to the lack of oxygen (Jones, 1933 ; Dore et 
al.,1959; Kordan, 1974).     
      In rice, the rate or percentage of germination is not influenced by O2 level and 
the coleoptile can develop and elongate under the reduced partial pressure of oxygen. 
However, root growth is poor and the growth of the primary leaf is barely detectable in an 
anaerobic condition (Nagai, 1916; Sasaki, 1926; Kordan, 1977; Takahashi, 1978 ; 
Turner et al.,1981).       
      Turner et al. (1981) and Won et al. (1999) suggested that the tolerance to low O2 
levels during the seedling stages varies with the cultivar. In water-seeding cultivation, 
the failure of establishment of rice seedlings has been resulted from the lack of O2. 
Therefore, it is necessary to screen or develop cultivars which are tolerant to low 
dissolved oxygen (DO) concentrations for water-seeded rice.
      Thus, in the present study we tried to screen several promising rice cultivars, 
which may be tolerant to low DO levels during the seedling stage especially in root 
growth for further breeding of water-seeded rice.     
     
Materials and Methods
      In the previous study, the seedling traits were not different among genotypes at 
0-0.5 and 1.0-1.5 mg L-1 DO levels except for coleoptile length. However, at 2.0-2.5 mg 
L-1 DO level, all the seedling traits were different among the cultivars (Won et al., 1999). 
Therefore, we carried out this study at the range of 1.5-2.5 mg L-1 DO concentration. The 
DO concentration was controlled using N2 gas which was bubbled through distilled 
water container, and every day the DO concentration was checked and controlled within 
1.5-2.5 mg L-1. A model ODT-101 O2/DO meter of SIBATA company was used to 
continuously monitor the DO level. 
      Seventy genotypes, 50 from Japan, three from Korea, three from IRRI, seven 
from the U.S., four selected lines (Won et al., 1998) and one each from Portugal, the 
Philippines and Hungary were used. Rice seeds were sterilized with 70% ethanol for 30 
second and 2% sodium hypochloride for 15 minutes, and then rinsed three times with 
water. The seeds were incubated for 2 days at 25 ℃ for uniformed presprouting. Ten 
presprouted seeds per each genotype were placed into the six seedbeds which were made 
of transparent acrylic plate with 6 x 12 x 7 cm depth and 12 partitions per one seedbed. 
Then, these seedbeds were submerged into oxygen-controlled water in boxes 7 cm in 
depth. The water boxes containing the seedbed were placed in a growth chamber, which 
provided 25℃ air temperature under 400 μmol m-2 s-1 light for 12 hours per day.        
      At eight days after placing seeds in the seedbed, the root number (longer than 5 
mm), the length of longest root, the length of coleoptile and the length of shoot were 
measured for five randomly selected seedlings from each cultivar. All treatments were 
carried out with two replications. Analysis of variance (ANOVA) and simple linear 
correlations among the seedling traits of this study were done using IRRISTAT program. 

Results and Discussion 
      The effects of genotypes were highly significant (p < 0.01) for all of the seedling 
traits such as root number, longest root length, coleoptile length and shoot length (Table 
1). This result suggests that there are genotype differences in tolerance to low DO levels 
during the seedling stages varies with the genotype and that the screening of the cultivars 
tolerant to low DO conditions is possible.
      Fig. 1 shows the relationship among the seedling traits under low DO. The range 
of the seedling traits of 70 genotypes were 0 - 5.1 roots per seedling in root number, 0 - 
42.6 mm in longest root length, 0 - 32.3 mm in coleoptile length and 0 - 112.9 mm in 
shoot length. The mean of 70 genotypes were 3.2, 23.6 mm, 25.1 mm and 58.1 mm for 
root number per seedling, longest root length, coleoptile length and shoot length, 
respectively. Generally, cultivars from IRRI and U.S. (mostly Indica type) are not 
desirable for the seedling growth under low DO condition compared with those from 
Japan and others (mostly Japonica type).  
      Root number was significantly correlated with coleoptile length and shoot length 
(r= 0.352** and r= 0.655**, respectively). Root length was also significantly correlated 
with coleoptile length and shoot length (r= 0.392** and r= 0.658**, respectively). Even 
though the correlation coefficients of the coleoptile length with root number or longest 
root length were significant, the values were lower than those of shoot length with root 
number and longest root length. This result somewhat agreed with the previous report 
that, at a lower DO levels, the initiation and elongation pattern of the root was the same 
as that of shoot but not as that of coleoptile (Won et al., 1999). 
      Arroz da Terra from Portugal was included in the best ten cultivars for all of the 
seedling traits (Table 2). Shinriki and Italica Livorno were included in the best ten for all 
of the seedling traits except for the root number. Therefore, Arroz da Terra, Shinriki and 
Italica Livorno were considered as the most tolerant to the low DO condition in the 
present study. Inoue et al. (1997) reported that Arroz da Terra showed a high rate of 
seedling establishment due to the low percentage of floating seedlings and seedling 
mortality. Kinmaze, Nakateshinsenbon, Takara and Norin 18 were included in the best 
ten cultivars for root number and longest root length, suggesting that they had relatively 
good characteristics in root trait. On the contrary, Yumetsukushi and Kinuhikari had 
good characteristics in shoot traits. The cultivars of Benisengoku, Cawwa/Fortuna 6-
103-15, Kamenoo and Line 53 showed the somewhat strong tolerance to both root and 
shoot traits.
      Many researchers considered the tolerance to low DO condition as one of the 
main factors necessary for the good establishment of the seedlings in the water-seeded 
rice. Chapman et al. (1963) reported that the concentration of dissolved oxygen in the 
soil of the germination zone had been depleted to less than 2 mg L-1 within 11 hours after 
sowing (14 hours after flooding), and it had been reduced to zero at 38 hours after 
flooding. In the present study, the range of the dissolved oxygen concentration was 1.5-
2.5 mg L-1, and there were wide varietal differences in root traits and shoot traits. 
Therefore, the promising cultivars screened in this study might be useful genetic sources 
in improving the establishment of the rice seedlings for water-seeding.                   

References
Chapman, A.L. et al. 1963. Crop Sci. 3:392−397.
Dore, J. et al. 1959. Trop. Agric. Trin. 36:15−34.
Inoue, N. et al. 1997. Jpn. J. Crop Sci. 66:632−639.
Jones, J.W. et al. 1933. J. Am. Soc. Agron. 25:69−81.
Kordan, H.A. 1974. New Phytol. 73:695−697.
Kordan, H.A. 1977. Plant Sci. Lett. 9:53−56.
Nagai, I. 1916. J. Col. Agric., Imp. Univ. Tokyo 3:109−158.
Sasaki, T. 1926. Rep. Agric. Sci. Soc. 288:461−469*.
Takahashi, N. 1978. Aust. J. Plant Physiol. 5:511−517.
Turner, F.T. et al. 1981. Agron. J. 73:566−570.
Won, J. G. et al. 1998. Plant Prod. Sci. 1:280−285.
Won, J. G. et al. 1999. Jpn. J. Crop Sci. 68. (Extra issue 2):222−223.
* In Japanese with English abstract.

Table 1. Mean squares of analysis of variance for root number, longest root length, 
   coleoptile length and shoot length for genotypes under the low dissolved oxygen level
   (1.5-2.5 mg L-1).

1)  Root number longer than 5 mm.
** : Significant at the 1% level of probability.

Table 2. The best ten cultivars for root number, longest root length, coleoptile length and 
   shoot length among 70 cultivars under low dissolved oxygen level (1.5-2.5 mg
   L-1).

1)  Nakateshinsenbon                   2)  Arroz da Terra(Portugal)   
3)  Cawwa/Fortuna 6-103-15 (Philippine)   4)  Italica Livorno (Hungary)
5)  Line 53 (Selected line)

Fig. 1. Relationships among the seedling traits under low dissolved oxygen condition
    (1.5-2.5 mg L-1).
    n= 70; AT: Arroz da Terra, IL: Italica Livorno, SR: Shinriki

以上