Larvae of Maculinea Footnote 1 alcon ([Denis & Schiffermüller], 1775) (Lepidoptera: Lycaenidae) start their life feeding on the developing seeds of gentian plants and in their last (fourth) instar they become social parasites of Myrmica Latreille, 1804 (Hymenoptera: Formicidae) ant colonies (see details in Thomas et al. 1989). The host ant species often vary in different regions (Elmes et al. 1998; Als et al. 2002; Höttinger et al. 2003; Tartally et al. 2008; Sielezniew and Stankiewicz-Fiedurek 2009). Knowledge of the local host ant species is important for the successful protection of these endangered butterflies and for answering evolutionary questions (e.g. Settele et al. 2005; Settele and Kühn 2009). As far as we are aware, no data on the host ant specificity of the Portuguese M. alcon populations has been published, and our aim was to fill this gap.

Myrmica nests were searched within 2 m (the approximate foraging zone of Myrmica workers, see Elmes et al. 1998) around Gentiana pneumonanthe L. host plants in three patches of a marshy mountain meadow with a lot of Erica tetralix L. plants in NE-Portugal. The meadow is near the village of Lamas de Olo (41°22′N, 7°48′W, 1,029 m a.s.l.) in the Alvão Natural Park (PNAl) and is part of a Site of Community Importance (SCI) for the Mediterranean biogeographical region, listed in 2006 by the European Commission (2006/613/EC of 19 July 2006). The population size of M. alcon was determined there as between 1,000 and 2,000 in 2009 (see more details about the site in Soares et al. 2010). The research was carried out in late June-early July, just before the flying period of the butterfly, in 2002 and 2007–2009. Myrmica nests found were carefully opened and searched for fully-grown larvae and pupae (“specimens” in the following) of M. alcon. After excavation, the ground and vegetation were restored to as close to their original condition as possible. We tried to find a compromise between getting enough data and not disturbing the small site of this endangered butterfly, so we spread the collection of data over several years. Five to ten workers were collected from each ant nest and preserved in 75% ethanol for identification in the laboratory (det. 2002: G.W. Elmes, 2007–2009: AT and SC according to Seifert 1988). Host ant specificity was calculated in two ways (see details in Tartally et al. 2008): p1 is the 2-tailed probability from a Fisher exact test of heterogeneity in infection of host ant nests (as implemented at http://www.quantitativeskills.com/sisa/), and p2 is the probability from a randomization test of ant nests between species (using Nash 2009).

We found M. aloba Forel, 1909 nests in all three patches investigated. Only one other Myrmca species, M. ruginodis Nylander, 1846 was found, and only in two patches. A total of 30 nests of the two Myrmica species were opened (M. aloba, n = 26; and M. ruginodis, n = 4), and 40 M. alcon specimens were found in 12 M. aloba nests. Thus, 46% of M. aloba nests were infested, with a mean number of M. alcon specimens in all nests of 1.54, and 3.33 in only the infested nests (range: 2–7). Although only M. aloba nests were infected, the small number of M. ruginodis nests found meant that we could not rule out this species as a potential host, and that measures of local host specificity were non-significant (p1 = 0.130, p2 = 0.190). The number of specimens per Myrmica nest was higher than one, and thus similar to other populations in Europe (Als et al. 2002; Tartally et al. 2008; Witek et al. 2008).

To the best of our knowledge, these are the first records of the host ant usage of any Maculinea species in Portugal, and of any Maculinea from M. aloba nests (for a review see Fiedler 2006; Pech et al. 2007). M. aloba was found as host in all three patches and in all four summers. Thus, this ant species appears to be the most important host of this butterfly in this locality. This is supported by the fact that only this Myrmica species was found in two other nearby M. alcon sites, although only five uninfested nests were found in total within 2 m of G. pneumonanthe (at PNAl near to Lamas de Olo: 41°21′N, 7°47′W, 1,000 m a.s.l., two nests; and 41°22′N, 7°48′W, 980 m a.s.l, three nests). However, care should be taken when generalising from the known host ant of a population to the host ant usage within a region (see e.g. Als et al. 2002; Tartally et al. 2008), therefore the host ant usage on the latter sites requires more investigation.

The distribution of M. aloba is restricted to the Iberian Peninsula and some nearby islands in Europe (Seifert 1988), typically on salt marshes with a high water table and an altitude of up to 1,400 m, and it is therefore not surprising that this ant has not been recorded as a Maculinea host from other regions. However, Elmes et al. (1998) report M. aloba from some Spanish M. alcon sites, but they found only Myrmica scabrinodis Nylander, 1846 as the host of M. alcon there. Myrmica ruginodis, the other recorded potential host ant, is known as an important M. alcon host in the Netherlands and Denmark, but not in other parts of Europe, despite its occurrence on some M. alcon sites there (Elmes et al. 1998; Als et al. 2002; Höttinger et al. 2003; Tartally et al. 2008; Witek et al. 2008; Sielezniew and Stankiewicz-Fiedurek 2009). Whether M. ruginodis is used by the Portuguese M. alcon populations is still an open question. Some other Myrmica species are also known as hosts of M. alcon in different parts of Europe: M. scabrinodis in Austria, France, the Netherlands, Hungary, Spain, Poland, Ukraine and Transylvania (Romania); M. vandeli Bondroit, 1920 in Hungary, Poland and Transylvania; M. rubra (Linnaeus, 1758) in Denmark, the Netherlands and Sweden and M. salina Ruzsky, 1905 in Hungary (see references above). All of these records are for the ‘classical form’ of M. alcon occurring on boggy meadows, wet heaths and fens. The host ant usage of the form inhabiting nutrient-poor xerothermic and calcareous mountain grasslands, usually called M. ‘rebeli’ (Hirschke 1904), is more complex (see details in Tartally et al. 2008).

Our results confirm that peripheral populations often adapt to special circumstances (e.g. Bernays and Graham 1988; Martin and Pullin 2004), in this case to a host ant that is unknown from other regions. According to these and to earlier results on different host ant species in different regions (see above), the host ant usage at the local scale needs to be considered for all European populations in the design and implementation of conservation management aimed to protect threatened Maculinea van Eecke, 1915 butterflies from local extinction, and especially in the case of peripheral populations. Future research could concentrate on clarifying the effect of the grazing regime on the host ants and plants at the Alcon Blue sites at Lamas de Olo, because according to our opinion it can be critical for the conservation of this butterfly there.