Abstract: Southern rice black- streaked dwarf virus disease was an important fresh viral disease in the southern China recently. In order to clarify the relationship between meteorological factors and the incidence and improve the prediction ability and avoid heavy losses, this paper studied their relationship based on the meteorological data and the rice black- streaked dwarf virus disease occurring data by stepwise regression analysis and path analysis. After selecting the factors, we established a regression model: y? =- 531.9999- 4.4850x1+1.0238x2+3.6891x3+7.0837x4+1.0286x5-2.8431x6. It showed that the regression prediction model matched effectively and had a strong forecast ability. This provided reference not only for Huazhou, but also for western Guangdong region.