Abstract
A method to demonstrate varietal differences in seedling establishment of rice under low temperatures and methods of data analyses are discussed. Nursery boxes were filled with sieved soil up to 0, 4, and 8 mm below the top. After seeding on the surface of the soil, soil was filled to the top to attain different seeding depths. The boxes were kept submerged in a paddy for 30 days (soil temperature was 11.3 to 20.9°C). The optimum depth for demonstrating the varietal difference was determined as 4 mm. The varietal difference observed with this method was almost identical to that observed in the experiments with the seeds directly sown in the paddy fields. From the results of the experiments with 4-mm deep seeding for five years, five varieties, Arroz da Terra, Calrose, Haenuki, Fukuhibiki, and Blue Bonnet were chosen empirically as the varieties with excellent, good, fair, poor and very poor seedling establishment, respectively. The power to detect the varietal differences was improved when analyses were made on the data exclusively obtained from experiments that establishment rate of Arroz da Terra was more than 75% and that of Fukuhibiki was in between 30 to 60%. Using the average and the standard deviation in these five standard varieties, quasi-deviation values for other varieties were simulated. Practically, the power to detect the varietal differences was improved by using the quasi-deviation values as compared with the average of establishment rate.
