Baseline toxicity bioassays

The bioassay technique used to determine differential susceptibility of F. occidentalis and T. tabaci to avermectin and beta-cypermethrin was similar to the adult bioassay methods described by Wang et al.⁴⁶. Only recently eclosed (1 day post eclosion) adult females were used in bioassays. Each insecticide was serially diluted to 6–7 concentrations with distilled water containing 0.1% Triton X-100 (Beijing Solar Bio Science and Technology Co. Ltd., China). Commercial formulations of avermectin and beta-cypermethrin insecticides (Hebei Veyong Bio-Chemical) were used in all trials. Leaves were cut from purple cabbage plants grown in a greenhouse without any insecticides, and then were dipped in the appropriate insecticide concentration for 10 s. Control leaves were treated in the same manner with a 0.1% Triton X-100 solution. Leaves were allowed to dry at room temperature for approximately 1 hour.

Modified ventilated glass cells were used as the bioassay chambers. A glass cell consisted of a center glass plate (4 mm thick) with two holes (20 mm diameter) drilled through the plate. The center glass plate was covered by a solid top and a bottom glass plate. The three plates were kept together with strong rubber bands forming two cells. Two ventilation holes were drilled opposite each other in the sides of the plate into the cell. The holes were covered inside the cell by copper wire screen with a mesh size of 200 μm.

After air-drying, each leaf was placed with its adaxial surface downward onto filter paper. The leaf and filter paper were placed individually between the center and bottom glass of a plate. Thrips were then introduced into the cells with the experimental leaves. Thirty adult female thrips were moved into each cell with an aspirator. Three cells were prepared for each concentration (total 90 adult thrips for each concentration) and kept under the environmental conditions described above. Mortality was recorded 48 h later. Thrips unable to move and showing characteristic effects of insecticide intoxication were scored as dead.

The data from these insecticide toxicity assays were subjected to probit analysis using POLO-PC (LeOra Software, Berkeley, CA) after correcting for control mortality with Abbott’s formula⁴⁷. Lethal concentration values (LC) were compared based on their 95% confidence limits, with non-overlapping intervals used to establish significance. Relative susceptibility ratios for the two species were calculated by dividing the LC estimates of F. occidentalis by the corresponding LC estimates for T. tabaci.