Visually and olfactorily enhanced attractive devices for thrips management.

'Lure-and-infect' is an insect pest management strategy with high potential but so far there are few examples of its application. Using traps as surrogates for auto-dissemination devices, we tested the attractiveness to naturally occurring thrips (Thysanoptera: Thripidae) of three trap types differing in colour and structure, with and without the thrips lure methyl isonicotinate (MI), and sticky plate traps as a control. The aim was to find more effective traps that could be further developed into devices for auto-dissemination and lure-and-infect of thrips. The number of thrips captured varied substantially with trap type and the presence of the MI lure. We found a high visual response to a sticky 'white ruffle' trap (i.e., a 30-cm-long cylindrical outline of folded fabric), compared to a commonly used blue sticky plate trap (Bug-scan) as the control. This effect was seen both in a greenhouse with roses ( spp.), where we encountered western flower thrips, (Pergande), and in a grass field, where we encountered onion thrips, Lindeman, and New Zealand flower thrips, (Crawford). In the absence of MI, the white ruffle trap caught 7-22× more thrips than the control Bug-scan trap. A similarly designed blue ruffle trap and a modified Lynfield trap caught lower thrips numbers than the white ruffle and the control Bug-scan traps. Presence of MI substantially increased the captures of in all three trap types in the field (2.5-18×). In the greenhouse, without MI the white ruffle trap caught 3.5-14× more thrips than the Bug-scan, blue ruffle, or modified Lynfield traps. Presence of MI increased the captures of males and females in the Lynfield and blue ruffle traps (1.4-2.8×), but not in the white ruffle trap in the greenhouse (ca. 1.1×). The importance of visual and olfactory factors for the design of effective auto-dissemination and lure-and-infect strategies for thrips management is discussed.