Melampyrum sylvaticum as a pre-diapause host plant of the scarce fritillary (Euphydryas maturna) in Finland

Abstract Background The scarce fritillary Euphydryas (Hypodryas) maturna (L.) is included in the Habitats Directive's Annexes II and IV(a). Therefore, it is crucially important to be able to define the habitat and breeding places of E. maturna in a correct and unbiased way. New information Data on a previously unknown pre-diapause main host plant, the small cow-wheat (Melampyrum sylvaticum L.), of Euphydryas maturna in Finland is presented.


Introduction
The scarce fritillary Euphydryas (Hypodryas) maturna (Linnaeus, 1758) is a high-profile species within the European Union, as it has been included in the Habitats Directive's (Council Directive 92/43/EEC of 21 May 1992 on the conservation of natural habitats and of wild fauna and flora) Annexes II and IV(a). Based on the Annex II, special conservation areas (i.e. Natura 2000 areas) need to be designated for E. maturna. The Annex IV lists species in need of strict protection, and those species and their breeding and resting places are protected by national legislation, which also applies to Finland. Therefore, the ability to define the habitat and breeding places of E. maturna in a correct and unbiased way is crucially important for both protecting the species effectively and not making uninformed administrative decisions which may be economically very costly. Euphydryas maturna is a wide-spread species in SW Finland, and it has been assessed as Least Concern by the IUCN criteria in Finland (Kaitila et al. 2010).
The species of the tribe Melitaeini, to which E. maturna belongs, feed mainly on plants containing secondary plant metabolites called iridoids (Bowers 1983, Wahlberg 2001. Iridoids are used for oviposition-plant selection, and as feeding stimulants and defensive chemicals by larvae (e.g. Bowers 1983, Nieminen et al. 2003, Wahlberg 2001. Larval host plants are the key part for the definition of breeding habitat for specialized herbivores such as E. maturna. Numerous plant species have been recorded as pre-diapause (Table 1) and/or post-diapause hosts of E. maturna throughout its range (see e.g. Wahlberg 1998, Dolek et al. 2013. In Finland, the common cow-wheat (Melampyrum pratense L.) has been recorded as the main host plant (Wahlberg 1998). Here, I present data on a previously unknown pre-diapause main host plant, the small cow-wheat (Melampyrum sylvaticum L.), of Euphydryas maturna in Finland.

Ecology
Totally 167 larval groups were located, all on Melampyrum spp. (Fig. 1). In some cases, at least two original larval groups had probably merged. All larval groups were either in clear-cuts (usually close to the edges, and sometimes within the forest 0-5 m from the clear-cut [Figs 2,4]), in thinned and light commercial forests (Figs 3, 4), or in open powerline corridors (Fig. 5).
Due to the dry conditions in July and August, many or even all host plants had  Clear-cut edge habitat of Euphydryas maturna. Clear-cut edges typically remain suitable for breeding for some years only until they become overgrown by tall grasses and tree seedlings.  Euphydryas maturna habitat in a commercial, thinned pine-dominated forest with ca. 30-year old trees, and in a clear-cut edge. This kind of forest habitat is probably suitable after thinning for several years, but longer than spruce-dominated forests (Fig. 3). Also, edge habitats in these relatively dry habitats overgrow somewhat slower than in moister edges (Fig. 2).

Discussion
The regional host plant use of E. maturna is highly variable throughout its range (Table 1), but is apparently restricted to only a couple of preferred species used for oviposition within any particular region. For example, Fraxinus is the most regularly used oviposition-plant genus in the Central Europe (e.g. Cizek and , Levente 2005, Freese et al. 2006, Dolek et al. 2013, whereas lower plants such as Veronica longifolia are often used in the eastern areas (e.g. Korshunov andGorbunov 1995, Gorbunov andKosterin 2007). However, it seems common that the post-diapause larvae feed on a wider spectrum of host plants than are used for oviposition (Gorbunov andKosterin 2007, Dolek et al. 2013), for example Plantago lanceolata is frequently used after diapause in Austria (Freese et al. 2006). There may be regional differences in preference also within the Finnish range, as all females observed during their search for oviposition-plants ignored M. sylvaticum in a study performed about 200 km to the northeast of this study area (Wahlberg 1998). Moreover, the importance of other host plants than M. pratense and M. sylvaticum still remain uncertain throughout Finland.
The use of M. sylvaticum as a host plant (Fig. 1) considerably increases both the suitable breeding area of E. maturna and the amount of resources available for it. In the study area, the increase in both of these variables must be manyfold, but remains to be quantified. Melampyrum pratense is much more vulnerable to desiccation and withering because it grows in drier sites than M. sylvaticum. The ability to use both of these Melampyrum species is extremely important in dry summers such as 2014, when more than 90% of host plants had withered in several sites. That high rate of dry host plants has likely increased mortality of groups of small larvae and may also decrease overwintering success due to starvation of larvae, which are common phenomena in another larval group-forming species Melitaea cinxia in Finland (e.g. Kuussaari et al. 2004).
Some leaves of Vaccinium myrtillus had been eaten within some larval webs. Even though I could not confirm that E. maturna larvae had eaten them, it is, however, likely because other herbivorous larvae were not observed and Melampyrum individuals were almost completely dry in and around these larval nests. Therefore, larvae may have used V. myrtillus to rescue themselves from starvation. The same explanation may apply to the odd observations of larvae feeding on e.g. Fagus, Populus and Salix regularly referred to in the literature (e.g. Korshunov and Gorbunov 1995, Gorbunov and Kosterin 2007, Dolek et al. 2013. A further explanation for 'odd' host records is that the actual host plants often become consumed completely leaving only non-host plants visible among and next to larval webs.