Electrostatic pollination by butterflies and moths

Abstract

Animals, most notably insects, generally seem to accumulate electrostatic charge in nature. These electrostatic charges will exert forces on other charges in these animals' environments and therefore have the potential to attract or repel other objects, for example, pollen from flowers. Here, we show that butterflies and moths (Lepidoptera) accumulate electrostatic charge while in flight. Then, using finite element analysis, we demonstrate that when within millimetres of a flower, the electrostatic charge of a lepidopteran generates an electric field in excess of 5 kV m^-1, and that an electric field of this magnitude is sufficient to elicit contactless pollen transfer from flowers across air gaps onto the body of a butterfly or moth. Furthermore, we see that phylogenetic variations exist in the magnitude and polarity of net charge between different species and families and Lepidoptera. These phylogenetic variations in electrostatic charging correlate with morphological, biogeographical and ecological differences between different clades. Such correlations with biogeographical and ecological differences may reflect evolutionary adaptations towards maximizing or minimizing charge accumulation, in relation to pollination, predation and parasitism, and thus we introduce the idea that electrostatic charging may be a trait upon which evolution can act.

Keywords: Lepidoptera; electric fields; flowers; insects; pollen; static charge.

Figure 1.

Macrophotographs of 5 of 11 lepidopteran species investigated within this study. (a) European peacock butterfly (Aglais io). (b) Postman butterfly (Heliconius melpomene). (c) Elephant hawk moth (Deilephila elpenor). (d) Male Ligurian emperor moth (Saturnia pavoniella). (e) Male Io moth (Automeris io).

Figure 2.

Example measurements of the variables r _L, r _W and S _All, as introduced and required for the equation derived for the surface area of a lepidopteran, as well as w _T, w _B and h, used for building a scale interpretation of a butterfly for finite element analysis. This example is applied to a peacock butterfly (Aglais io).

Figure 3.

(a) A typical example of the voltage generated by the picoammeter over time as a charged lepidopteran flew freely through the ring electrode system. An offset has been applied to bring the baseline to 0, and a notch filter between 49 and 51 Hz to reduce 50 Hz noise from mains electricity in the United Kingdom. This particular example was a peacock butterfly (Aglais io) carrying a net electrostatic charge of +15.41 pC. (b) Distribution of net electrostatic charges carried by free-flying Aglais io butterflies. Mean ± s.d. = +49.54 ± 57.22 pC, median = 39.40 pC, n = 72.

Figure 4.

(a) Three-dimensional computational model of the electric field formed between a typically charged lepidopteran and a grounded flower. Colour scale represents electric field strength, with data truncated above 5 kV m⁻¹ for clarity. Grey indicates model geometry. (b) Three-dimensional computational model simulating pollen grain trajectories (n = 100) under the influence of electric, gravitational and drag forces, for a typically charged lepidopteran positioned 6 mm from the stamen of a flower. Blue circles show the final locations of individual pollen grains. Colour scale represents the velocity of pollen grains. All models are produced using the finite element method.

Figure 5.

Different characteristics of the net electrostatic charge carried by various species of butterfly and moth. Conifer trees denote temperate species, palm trees denote tropical species, flowers denote floral visitors, flowers crossed-through denote species that do not visit flowers, suns denote diurnal species and moons denote nocturnal species. The presence of both a sun and moon denotes that intraspecies differences in the activity regime exist between sexes, with the females’ activity regime indicated by the top left icon, and the males’ activity regime indicated by the bottom right icon. Points indicate individual measurements; box and whisker show the median, lower and upper quartiles and range; half-violin shows the distribution of each dataset. (a) The net electrostatic charge. (b) The net electrostatic charge magnitude. (c) The net electrostatic charge density is calculated by normalizing the net electrostatic charge by an estimation of each species’ surface area.

Figure 6.

Predicted values based upon surface area (a) and native climate (b), from the linear mixed-effects model for the transformed net electrostatic charge magnitude, with surface area and native climate as fixed effects, and day of measurement, species and family as random effects. Predicted values based upon ecological guild (c) and native climate (d), from the generalized linear mixed-effects model for the net electrostatic charge polarity, with ecological guild and native climate as fixed effects, and day of measurement and species as random effects. Blue denotes predicted values, lines or the shaded area denote 95% confidence intervals. Asterisks indicate statistically significant differences: * indicates p < 0.05, ** indicates p < 0.01.