Figure 1. Effect of imidacloprid on arthropod communities and spider mite abundance in New York and Maryland.

(A) Abundance of arthropods (per cm² of leaf area) on imidacloprid-treated elms (N = 10) and on untreated trees (N = 10). Asterisks mark differences in overall abundance of arthropods that were significant within each year (P<0.05; Monte Carlo permutation test). At both locations, arthropod communities increased on elms that received imidacloprid. Abundance of spider mites, Tetranychidae, explained most of the variation due to imidacloprid treatments. Pie charts represent percent contribution of the most abundant taxa to the sampled arthropod community over three-year period at each location. (B) Abundance (√(number)/cm²) of the spider mite, T. schoenei, on elms treated with imidacloprid (N = 10) and on untreated trees (N = 10) in New York and Maryland. Asterisks mark means±s.e.m. that differed significantly (P<0.05). There was a significant interactive effect of treatment and time for both locations and in most years, and means were compared within each date (Table S1). Elevated densities of mites were found only on elms treated with imidacloprid. Rarely encountered taxa included arthropods in families Chrysopidae, Coccinellidae, Cecidomyiidae, Aphididae, Saproglyphidae and Thripidae. These arthropods were pooled and categorized as ‘Other’.