Learning Of Repellents Derived From Non-host Plants By A Specialist Herbivore

Plant-originated repellents can repel insect pests and deter them from laying eggs. As such repellents may prevent plants from attack by insect pests before their arrival, they have received particular attention in the field of pest control. In this study, we investigated the role of learning in affecting the effects of non-host plant-originated repellents, using as a model system the specialist herbivore, diamondback moth (Plutella xylostelld) and two kinds of repellents, a methanol extract of Chrysanthemum morifolium and a commercial insect repellent BIOACT-T35 which has azadirachtin from neem trees as its main active ingredient.(1) Both kinds of repellents tested repelled ovipositing diamondback moth females. When untreated (sprayed with distilled water) plants of Chinese cabbage and plants sprayed with either a chrysanthemum extract at a rate of 100g/L or BIO ACT at a rate of 10ml/L were presented together in choice tests to diamondback moth females, the number of eggs laid on untreated plants was 7.5 or 10.6 times higher than that on treated plants. Further tests with the chrysanthemum extract indicated that effects of repellence were positively correlated with concentration of the repellent.(2) Diamondback moth expressed habituation to and induced preference for non-host originated repellents. When diamondback moth pupae were immersed in a solution of chrysanthemum extract for 10 seconds, or female moths were exposed to residue of the solution for 10 minutes after emergence, or the insects were exposed to the repellent at both pupal stage and early adulthood, the females subsequently spentlonger time in the arm connected to a plant treated with chrysanthemum extract than in the arm connected to a untreated plant in Y-tube olfactometer tests. These experienced females also deposited higher number of eggs on plants treated with chrysanthemum extract than on untreated plants. These results demonstrated substantial habituation to or even induced preference for the repellent by diamondback moth, making the plant extract non-repellent or even attractive to the insects. Further tests with plants unevenly sprayed with this repellent showed that when females have encountered alternatively plant surface with the repellent and that without the repellent, they became less sensitive to the repellent.(3) Responses of habituation and induced preference would change with the concentration and formulation of repellents. When either the number of times of experience or the duration of experience with 10ml/L BIO ACT by the female moths increased, the insects increased the proportion of eggs laid on plants sprayed with this repellent in choice tests using untreated and treated plants. However, no induction of preference for BIO ACT was observed. When a higher concentration (15ml/L) of BIOACT or a mixture of BIOACT and an ethanol extract of orange peels was applied onto the host plants, no habituation by the insets was found.(4) Effect of repellents decreased on plants with higher intrinsic attractiveness. Barbarea vulgaris has much stronger intrinsic attractiveness to diamondback moth than common cabbage. When either common cabbage plants sprayed with repellents or plants of B. vulgaris sprayed with repellents and untreated plants of the same species were exposed to females in choice tests, the insects were repelled to a similar extent by the treated plants in the two cases. However, when treated plants of common cabbage and treated plants of B. vulgarisi were exposed to the females in choice tests, the insects showed strong preference for the latter speciesof plants.(5) Volatiles of the Chrysanthemum morifolium were collected by circumfluence extraction and analysed by GC-MS. In all 27 components were identified, including mainly monoterpenoids, sesquiterpenoids and oxygen derivatives.Our study represents the first report of habituation and induced preference for repellents originated from non-host plants by a specialist herbivore, and demonstrates that such behaviour o...