In order to determine the effects of the plant odours in the tea plantation ecosystems on the adult behaviour of the tea geometrid Ectropis obliqua (Prout), the fresh tender shoots of seven species of common plants in the tea plantation community were tested as odor sources at the dosages of 1.5, 5.0, 7.5, 10.0, 15.0, 20.0 and 30.0 g, respectively, against the clear air as CK1 using a Y-shape olfactometer. The results showed as follows: ① The number of the geometrid adults responding to peppermint Mentha haplocalyx was significantly more than that to CK1 at the dosage of 1.5 g (P< 005), the number descended in a non-monotonic parabolic curve fashion with dosage, and was even significantly less than that to CK1 at the dosage of 30.0 g (P<0.05). ② The responses to rosemary Rosmarinus officinalis, tea Camellia sinensis and lemon balm Melissa officinalis increased in a non-monotonic “inverted U-shaped” dose-response curve pattern; and the numbers of adults responding to these three species of hosts were significantly more than those to CK1 (P＜0.05) at the dosage of 7.5 g, 10.0/15.0 g, and 10.0 g, respectively. However, they exhibited repellent effect on the adults at higher dosages. The number of the geometrid adults responding to rosemary was significantly less than that to CK1 at the dosage of 30.0 g (P<0.05). ③ Host marigold Tagetes erecta displayed a weak attraction at the lowest dosage (1.5 g), and showed a weak repellency at higher dosages. The number of the adults oriented to marigold was significantly less than that to CK1 at the dosage of 30.0 g (P<0.05). ④ Host patchouli Pelargonium odoratissimum showed no effects at the dosage of 5.0 g, but exhibited a weak attraction at the dosage of 7.5 g, and a weak repellent effect as the dosage further increased. ⑤ Non-host lavender Lavandula officinalis was inactive at the dosages of 5.0, 7.5 and 10.0 g, but exhibited a weak repellent effect at higher dosages. The number of the geometrid adults responded was significantly less than that to CK1 at the dosage of 30.0 g (P<0.05). Hexane extracts of fresh tender shoots of peppermint, marigold and tea were loaded onto rubber septa; they were immersed in the plant extracts for 24 hours before testing. Five types of rubber septa lures were prepared, i. e. 1.5 g peppermint equivalent/lure, 0.75 g peppermint equivalent/lure, 1.5 g marigold equivalent/lure, and 1.5 g tea equivalent/lure plus a hexane immersed lures as CK2. Each lure was attached onto an earthy yellow sticky plate for a seven day field bioassay in tea plantations during the adult stage of the 5th generation. These yellow plates were checked and replaced every day. All the treatment lures and CK2 exhibited a strong attraction to the moth.Traps baited with 1.5 g peppermint-lure caught significantly less moths than did the CK2, whereas other plant extract lures showed significantly higher trap catches than did the CK2 (i.e. 1.5 g tea/lure >0.75 g peppermint/lure > 1.5 g marigold/lure > CK2). During the larval stage of 5th generation, three trapping plots and three check plots (CK3), about 1/6 hectare each were selected and the larval population density in each plot was investigated. During the emergence of the 5th generation adults, seven sets of the above-mentioned five types of traps were placed in each trapping plot for seven days. And these yellow plates (traps) were checked and replaced every day. Compared with the 5th genetration, the revised larval population decreasing percentage of the 6th generation was 30%. Our results suggest that the plant odours might effectively affect the adult behaviour of the tea geometrid, with the peppermint being slightly more attractive than tea whereas the lavender being significantly repellent.