Evolutionary change in flight-to-light response in urban moths comes with changes in wing morphology

Moths and other insects are attracted by artificial light sources. This flight-to-light behaviour disrupts their general activity focused on finding essential habitat resources, such as mating partners, and increases predation risk. It thus has substantial fitness costs. In illuminated urban areas, spindle ermine moths Yponomeuta cagnagella were reported to have evolved a reduced flight-to-light response. Yet, the specific mechanism remained unknown, and was hypothesized to involve either changes in visual perception or general flight ability or overall mobility traits. Here, we test whether spindle ermine moths from urban and rural populations—with known differences in flight-to-light response—differ in flight-related morphological traits. Urban individuals were found to have on average smaller wings than rural moths, which in turn correlated with a lower probability of being attracted to an artificial light source. Our finding supports the reduced mobility hypothesis, which states that reduced mobility in urban areas is associated with specific morphological changes in the flight apparatus.

Methods
Caterpillars of the moth Yponomeuta cagnagella were collected from different populations (with multiple families per population) at either light-polluted, urban sites or pristine dark, rural areas. They were grown under common-garden conditions in the lab. Obtained adults received an individual code before they were subjected to an experimental test. In this test, moths were released at one side of a flight-cage and at the other end of the cage there was a Heath actinic (6 W) light trap. After 8h, it was recorded whether the moths were captured in the light trap or not (i.e., the flight-to-light response). Moths were stored in the freezer (-20 °C) after the experiment.

Thorax and abdomen mass – Dead adult moths were oven-dried at 60 °C for 4 h. Remaining antennae and legs were systematically removed in all individuals prior to measuring total dry body mass (mtot). Head, thorax, abdomen and wings were carefully separated. For each individual, thorax (mth) and abdomen mass (mab) were also measured. All mass measurements were done with a microbalance (precision ± 0.001 mg). From these data, relative thorax (mth,rel) (mth/mtot) and abdomen mass (mab,rel) (mab/mtot) were calculated.

Wing morphology - For each individual, we took standardized pictures of the ventral side (clear difference between wing surface and fringe) of the right forewing (Fig. 1). In case of significant wing damage, the left forewing was used instead. Forewing length (FWL), forewing width (FWW) and forewing area (A) were measured using ImageJ software (https://imagej.net/ij/index.html) on size-calibrated pictures. Length and width were measured twice and the mean value was used in the analyses. Based on these wing measures, we also calculated aspect ratio (AR) (4 ∙ FWL²/FWW) and wing loading (WL) (mtot/A).