Abstract
With a combination of biological, analytical, electrophysiological, and video-optical methods, it was possible to show that low concentrations of the new chloronicotinyl insecticide, imidacloprid, strongly affect the behaviour of Myzus persicae (Sulz.), leading eventually to the death of the aphids. Tests to elucidate the biological properties were performed under laboratory conditions with cabbage leaf petioles placed in insecticidal solutions over different periods of time. LC15(24h) values were considered as low concentrations and calculated for imidacloprid and pirimicarb, respectively. Imidacloprid at low concentrations depressed the honeydew excretion of apterous adults of M. persicae by almost 95% within 24 h without affecting the vitality of the majority of aphids, whereas, at equitoxic concentrations, pirimicarb showed much weaker effects on honeydew excretion, which strongly coincided with mortality. In choice experiments with alate morphs of M. persicae over 48 h, their larvae almost always occurred on the untreated control leaf, and were not found on the leaf which was treated systemically with low concentrations of imidacloprid. Apterous aphids placed on cabbage leaves systemically treated with low concentrations of imidacloprid showed nearly the same decrease in weight as untreated starving aphids, suggesting that their death was caused by starvation. Aphids that were moved from imidacloprid-treated to untreated leaves after 24 h began feeding on the latter and showed a steady increase in weight and honeydew production. This suggests that the behavioural response is reversible. Aphids on pirimicarb-treated (equitoxic dose) leaves showed no decrease in weight. Electrical penetration graphs revealed that M. persicae on artificial membranes containing imidacloprid probed more often before feeding than aphids on control sachets. Time-lapse videofilming of apterous adults placed on cabbage leaves revealed a migration from the leaf treated with low concentrations of imidacloprid to an untreated leaf. From the results of these experiments and the observed symptomatology it is possible to postulate two different and dose-dependent modes of action of imidacloprid on M. persicae: (1) the well-known mode of action with visually obvious irreversible symptoms (paralysis, tremor, uncoordinated leg-movement) at field rates, and (2) the reversible starvation response as an antifeedant effect, which is not coupled with typical symptoms of neuronal disorder, at lower concentrations.