Abstract: In order to reveal the effects of the main economic traits on yield of early indica rice, 29 varieties (lines) in early indica rice regional trials in Southern China from 2009 to 2011 were analyzed by? statistical methods such as correlation analysis, path analysis and principal component analysis. The results showed that: effective spikes number (12.78%) had the largest coefficient of variation in all the main economic traits; partial correlation analysis showed that the yield and effective panicle number (0.5632**), 1000 grain weight (0.7456**) showed a highly significant positive correlation; principal component analysis showed that the main factor of the early indica rice yield information was grains factors, the contribution rate of 39.93%, growth period factor, the contribution rate of 20.28%, grain weight factor, the contribution rate of 17.30%; the first three principal components cumulative contribution rate reached 77.51%. Path analysis showed that four yield components of direct path coefficient: total grains number (Py2 = 0.9601) > grain weight (Py4 = 0.7309) > seed setting rate (Py3 = 0.6872) > effective panicle number (Py1 = 0.5411). Regression analysis showed that the 72.56% yield was decided by effective panicle number, total grains number, seed setting rate, grain weight. Accordingly these results, the main direction of breeding and cultivation of early indica rice is presented.