Objective A mathematical evaluation model, D(x), was established to provide a reference for the evaluation, identification and genetic improvement of seed vitality traits in rice.

Method Taking 25 rice seeds as materials, multiple indexes were determined through standard germination experiment, tetrazolium staining identification, field simulation experiment and Q2 technology, and these indexes were analyzed by principal component analysis combined with membership function method and multiple linear regression, to establish the mathematical evaluation model for seed vitality, D(x). And the D(x) model was used to evaluate the influence of different cultivation densities on rice seed vitality.

Result The results showed that the average values of germination vigor, germination rate, 7-day bud length, hundred-seed fresh weight, field bud length, field root length, starting metabolism rate (SMR), increased motabolism time (IMT), relative germination time (RGT), and tetrazolium staining rate of indica rice were all higher than those of japonica rice. However, only the average values of 7-day root length, hundred-seed dry weight, and oxygen metabolism rate (OMR) of indica rice were slightly lower than those of japonica rice. In the Q2 test, japonica rice exhibited better performance in IMT value, RGT value, and other indicators. Correlation analysis of 13 rice seed vitality indexes revealed varying degrees of correlations among them. In the Q2 test, SMR value was extremely significantly correlated with IMT value, OMR value, RGT value, and hundred-seed dry weight, and significantly correlated with field bud length. OMR value was extremely significantly correlated with SMR value, IMT value, RGT value, and significantly correlated with germination vigor. A seed vitality comprehensive mathematical evaluation model Y(x) = 0.693+0.003X1-0.006X2-0.127X6-0.004X8+0.234X9 (R²=0.94) was established through stepwise regression. The predicted vitality values (Y) obtained from these five indexes were in close agreement with the order of field emergence rate among the tested materials, and they were highly correlated (r=0.83), indicating that this regression equation could effectively predict rice seed vitality.

Conclusion Five individual indexes, namely, germination vigor, germination rate, hundred-seed dry weight, oxygen metabolism rate (OMR) and germination initiation time (IMT) could be selected as the basis for the evaluation of seed vitality.