Colour preferences of reared and wild Cacopsylla pyri in relation to sex and infection with ‘Candidatus Phytoplasma pyri’

Comparing laboratory and field experiments is crucial to understand the ecological relevance of insect behaviour. In particular the use of reared and wild insects in behavioural experiments should be carefully considered and decided upon. In this study, we investigated colour preferences of the pear psyllid Cacopsylla pyri for different green colours, using both reared and wild-caught individuals. C. pyri is the main vector for ‘ Candidatus Phytoplasma pyri’, which is a phloem-limited bacterium that causes pear decline, a disease that damages the fruit and negatively affects the tree’s fitness. To investigate, whether the colour of non-symptomatic pear leaves may differ in relation to infection, spectral reflectance measurements were conducted, revealing no consistent differences. Based on these measurements, sticky traps were designed to match the natural reflectance of leaves in three different green colours. Laboratory experiments with reared C. pyri showed no clear preference for any of the three colours, though males exhibited a slight tendency towards light green (532 nm). In contrast, wild-caught psyllids tested under identical laboratory conditions strongly preferred light green traps, regardless of sex. Field experiments did not confirm these preferences. Non-target arthropods exhibited different responses: medium green (541 nm) and dark green (551 nm) traps captured higher numbers, highlighting that trap colour influences bycatch too. Over time, seasonal dynamics showed a moderate shift in psyllid preference towards medium green over time, whereas other arthropods increasingly preferred light green. Our findings demonstrate that the visual behaviour of psyllids is finely tuned to specific green colours, but that individuals reared in a laboratory may not fully represent behaviour of wild animals. The strong agreement between wild-caught laboratory assays and field experiments underscores the ecological validity of a green colour with a wavelength peak of 532 nm as an attractive trap colour. This emphasises the importance of validating laboratory results under natural conditions.