Abstract: The effects of tree species, crown size and insecticides on the absorption and conduction of insecticide applied by trunk injection has been investigated. The 2.8% emamectin benzoate + 0.3% carbosulfan (hereafter 3.1% EB), 6% emamectin benzoate + 4% imidacloprid (10% EB), and 2% emamectin benzoate + 0.5% abamectin (2.5% EB) was made from a mixture of emamectin benzoate with carbosulfan, imidacloprid and abamectin, respectively. The influences of tree species and insecticides (3.1% EB and 10% EB), and crown diameter (2.5% EB) on the absorption conduction of those insecticides in Citrus grandis were studied by flowing trunk injection, meanwhile the residual period of 2.5% EB was investigated by the high performance liquid chromatography (HPLC-UVD) method. The results showed that the absorption rate of insecticides was significantly affected by tree species: Salix babylonica > C. grandis > Koelreuteria paniculata. Moreover, the insecticides were completely absorbed into the trunk within 5 days in S. babylonica. However, the remaining 3.1% EB and 10% EB in K. paniculata were up to 66.6% and 48.6%, respectively. Although the absorption rate of 10% EB (12.6% remaining) in C. grandis was faster than that of 3.1% EB (29.3% remaining ) within 5 days, there was no significant difference between these two insecticides in other treatments. The absorption of 2.5% EB was positively related to the crown diameter of C. grandis. On the 113rd day after injection, abamectin and emamectin benzoate were still observed in the branches and leaves. The overall analysis suggested that the best absorption of 3.1% EB or 10% EB was observed in the case with S. babylonica. And there was no significant difference between the absorption of 3.1% EB and that of 10% EB among three tested trees. In addition, the absorption of 2.5% EB was positively related to the crown diameter of C. grandis, and the residual period of 2.5% EB in C. grandis was up to 90-120 days.