Phengaris (Maculinea) teleius butterflies select host plants close to Myrmica ants for oviposition, but P. nausithous do not

Abstract

Many lycaenid butterfly species have interactions with ants, with 12% obligatorily depending on two sequential sources of larval food, namely host plants and host ants. When host plants are abundant but the density of host ant nests is relatively low, most host plants have no host ant nest in their vicinity and are thus unsuitable for larval survival. Obligatorily myrmecophilous female butterflies, whose caterpillars feed on ant larvae, would have a comparative advantage if they deposit eggs on host plants in the proximity of a host ant nest. However, this ant‐mediated oviposition has been hotly debated. In an open‐air insectory experiment, we tested whether oviposition is ant‐mediated or not for two obligatory myrmecophilous butterfly species, Phengaris (Maculinea) nausithous Bergsträsser and Phengaris teleius Bergsträsser (Lepidoptera: Lycaenidae). Female butterflies could select host plants close to either no ant nest or a nest of one of the two Myrmica species (Hymenoptera: Formicidae) that are thought to be their host ant. Our results support ant‐mediated oviposition in P. teleius, but also indicate that there is no preference for either of the Myrmica species. More eggs were deposited and more caterpillars were found on long flowerheads close to Myrmica ant nests than on those without nests. Our findings suggest that ant presence is more important than flowerhead phenology for females of P. teleius. In contrast, P. nausithous females were not attracted by ants but preferred long flowerheads with a low percentage of green coloration. With these findings, this study contributes to better understanding of the habitat requirements of two highly specialized butterfly species.

Resource overlap and dilution effects shape host plant use in a myrmecophilous butterfly

“…Spatial overlap between the distribution of host plants and host ant nests at larger scales has previously been shown to increase the density and population persistence of the congeneric species P. arion (Casacci et al., ; Thomas, Simcox, & Clarke, ). Other studies with Phengaris butterflies have demonstrated effects of host plant phenology (Thomas & Elmes, ; Valdés & Ehrlén, , ), Myrmica presence (Van Dyck et al., ; Wynhoff & van Langevelde, ; Wynhoff et al., ) or both of these factors (Casacci et al., ) for oviposition site selection. Our study suggests that the spatial overlap between these two resources also at the scale of the host plant individual is associated with butterfly oviposition preference, and that the effects of plant phenology and ant abundance are interactive.…”

“…Previous observational studies have suggested that oviposition is host species dependent (Van Dyck et al., ). Experimental studies have found Phengaris oviposition to be dependent on Myrmica ant presence, but found no effect of species identity (Patricelli et al., ; Wynhoff & van Langevelde, ). Lastly, there are several studies that have found no evidence of ant‐dependent oviposition (Fürst & Nash, ; Nowicki, Witek, Skorka, & Woyciechowski, ; Thomas & Elmes, ).…”

“…early‐flowering). When butterflies approach host plants, or after alighting on a plant, females may use both visual cues to evaluate plant phenology and chemical cues to detect ant presence (Patricelli et al., ; Wynhoff & van Langevelde, ; Wynhoff et al., ), thereby evaluating the spatial overlap of the two resources needed by their offspring. If females use plant developmental stage as a cue indicating the availability of high‐quality food for the caterpillars, host plants surrounded by late‐flowering neighbours would be more attractive to butterflies than host plants surrounded by early‐flowering neighbours.…”

“…Caterpillars mimic the surface chemistry of the ant brood (Nash, Als, Maile, Jones, & Boomsma, ) and the acoustic signals of queen ants (Sala, Casacci, Balletto, Bonelli, & Barbero, ), and these features cause ants to carry them to their nest, where they spend the rest of their larval period as brood parasites. Females of Phengaris detect Myrmica presence through plant chemicals produced by ants or by plants in response to root tissue damage caused by ants (Patricelli et al., ; Van Dyck et al., ; Wynhoff & van Langevelde, ). Contrary to the majority of Phengaris species, which prey on ant brood, P. alcon is a “cuckoo” species (Als et al., ), and caterpillars feed primarily on regurgitants from ant workers, trophic eggs (i.e.…”

“…Moreover, the mobility of caterpillars after leaving their host plants is low (Hayes, ), and further development is possible only if the foraging range of suitable host ants is sufficiently close to the host plant individual (usually a few metres). Butterflies are therefore expected to oviposit on host plants in suitable development stages only if plants occur at micro‐sites where ant hosts are present (Wynhoff, Grutters, & van Langevelde, ; Wynhoff & van Langevelde, ).…”