The evaluation of spots in rice seeds has been performed through the calculation of the percentage of spotted seeds. However this measurement is not quantitative. The objective of this experiment was to compare the percentage of spotted seeds with an index of infection intensity, which gives a more precise quantitative information about the extension of the seed surface area infected. For this comparison, 5g-rice-seed samples from two field trials, carried out in Jaboticabal and Pindorama, in the State of São Paulo, Brazil, in 1982/83, were used. Each trial had the same 16 cultivars with four replications, totalizing 64 samples per site. Seeds of each sample were visually separated in one of four categories: n0 = spotless seeds; n1 = up to 5% of the seed surface spotted; n2 = from 6 to 25% of seed surface spotted; and n3 = over 25% of the seed surface spotted. The number of seeds was counted in each category in order to calculate the index of infection intensity. The same seed samples were separated in two groups: spotted and spotless seeds, independently of the surface area attacked. The number of seeds in each group was counted in order to calculate the percentage of spotted seeds. After the statistical analysis (F test and correlation) performed, the results showed highly significant correlation between both methods, and the so called "spot intensity index" presented better discrimination among cultivars than the percentage of spotted seeds. The authors suggest the use of the new index for selection of rice cultivars aiming to less spotted seeds.
rice; Oryza sativa L.; seed spots; methodology of seed spot evaluation; spot intensity index