A bee or not a bee: an experimental test of acoustic mimicry by hoverflies

2016-08-21T22:07:20Z (GMT) by Christopher Hassall
<p>The degree of similarity between Batesian mimics and their models varies widely and occurs across a range of sensory modalities. We use three complementary experimental paradigms to investigate acoustic mimicry in hoverflies (Diptera: Syrphidae) which mimic stinging Hymenoptera. First, we analyse sounds made by 13 hoverfly species and 9 Hymenoptera species with and without simulated predation (“alarm” and “flight” sounds, respectively). We demonstrate that the bumblebees <i>Bombus terrestris, B. hortorum, and B. lucorum,</i> and the hoverfly <i>Cheilosia illustrata</i> exhibit alarm sounds that are significantly different to their respective flight sounds, and indistinguishable between species. We then demonstrate that the <i>B. terrestris</i> alarm sound reduces predation on artificial prey by wild birds, but that the hoverfly mimic alarm sound does not. Finally, we trained chicks to avoid distasteful food in the presence of different acoustic stimuli. Overall the chicks showed no difference in response to bee and hoverfly stimuli, demonstrating no innate aversion to the <i>Bombus</i> alarm sound. We therefore conclude that (i) similarity of acoustic signals exists among Hymenoptera and hoverflies, (ii) acoustic aposematic signals (but not the almost identical mimetic signals) are effective at reducing predation, and (iii) wild birds exhibit learned rather than innate aversion to certain acoustic stimuli. </p>