Devillers, P. & Devillers-Terschuren, J.A. 2000. Characters and ecology of Epipactis pollinensis on Mount Pollino, its place in the constellation of E. viridiflora and E. pseudopurpurata. Extended summary. Royal Belgian Institute of Natural Sciences website, www.naturalsciences.net/cb. [Original published in French in Natural. Belges 81 (Orchid. 13)].
Epipactis pollinensis has recently been described from Mount Pollino, southern Italy (Baumann & Baumann 2000). Our observations, made in 1988, are presented as additional information. Sites are at altitudes between 1450 and 1530 metres in beech forests of Mount Pollino. The reduced height, inflorescence length, leaf size, leaf flower number (Table 1), the laxity of the inflorescence, compared to E. viridiflora confirm the diagnosis of Baumann & Baumann (2000). The link between hypochile and epichile, a character of the lip found useful in the definition of species within other groups of Epipactis (Devillers & Devillers-Terschuren 1999a, 1999b) also differs between the two species. In E. viridiflora, the walls of the hypochile progressively converge distally, giving the hypochile a drop-like shape tapering into a narrow groove that ascendds onto the epichile. In E. pollinensis, the rim of the hypochile is circular and connected to the epichile by a relatively wide channel with parallel, or even distally diverging, walls. E. pollinensis is, like E. viridiflora, a late-flowering species. Plants observed were either in bud or opening first flowers, while E. helleborine was in full- to end of flowering and E. gracilis at the end of flowering. The presence of numerous dipterids in the hypochile cup suggests that the plant is nectar-producing. The species is at least partly or largely allogamous, with functional viscidium, although a rapid drying of the rostellar gland together with pollinias in place, was observed.
The members of the E. viridiflora (E. purpurata) constellation are reviewed. Three Ponto-Hyrcanian species are frequently attached to the group. It appears, however, that E. bithynica belongs to the E. tremolsii constellation. E. condensata is difficult to evaluate. Its ornate lip and stem pilosity suggest a link with the E. tremolsii, or, more probably, the E. atrorubens group; its early flowering time, before that of E. helleborine, fits with this hypothesis. E. rechingeri, on the contrary, strongly resembles E. viridiflora and its close allies, differing from E. viridiflora and E. pollinensis by flowers often strongly tinged red, but resembling E. pollinensis by few and small leaves, and by a tendency to a lax inflorescence. Plant height, inflorescence height, large floral dimensions are similar to those of E. viridiflora (Table 2). E. halacsyi has flowers strongly tinged red as in E. rechingeri. Floral dimensions, number and dimensions of caulinar leaves and height of the plant seem to be near E. pollinensis. The inflorescence is lax, but the number of flowers seems to be greater than in E. pollinensis. E. pseudopurpurata was described by Mered'a (1996) from the Strazovske chain of the Slovakian southwestern Carpathians. On 7 August 1991, we found a colony of 7 plants with characters corresponding to this species growing in a beech forest in the Kiserleti Nature Reserve of the Bakony range, in western Hungary. This previously unpublished record is apparently the only one for Hungary. E. pseudopurpurata approaches E. pollinensis by several characters, in particular, the reduced height of the plant, its inflorescence and flowers. The leaves are even smaller and fewer than in the Italian species. Floral parts differ by a more slender column, a reduced rostellum, and crumbling pollinias, characters corresponding to habitual autogamy. The habit of producing single stems, instead of multi-stem clumps, apparently peculiar to the species, was confirmed in Hungary.
The beech forests of Mount Pollino belong to the complex of southern Italian beech forests of the Geranio versicoloris-Fagion (Bonin 1968; Gentile 1969; Bonin & Gamisans 1976; Pignatti 1998). The mid-altitude forests of this complex are grouped, in the most recent revision of Italian forests (Pignatti 1998), into a single entity corresponding to the Aquifolio-Fagetum association. They constitute, however, a very fragmented and insularised ensemble whose components can be distinguished, in particular, by the orchid cortège, which had incited us to define for them seven separate units in the CORINE-Biotopes classification (Devillers et al. 1991). E. pollinensis is presently known from only one of the islets of this beech forest archipelago, that of Mount Pollino, unit 41.183 of the CORINE-Biotopes classification. There are observations of Epipactis of the E. viridiflora constellation in the Apennines outside of the range of the southern Italian beech forests, or at its limits. The exact identity of these Epipactis has not been studied, but photos published by Conti & Pellegrini (1990) and Liverani (1991) seem, contrary to Baumann & Baumann's (2000) suggestion, to belong without doubt to E. viridiflora.
In the present state of knowledge, E. pollinensis should be considered as endemic to the beech forests of Mount Pollino, and as a rare species. We share Baumann & Baumann's (2000) opinion that the species is threatened. It appears to be, like its Hellenic vicariant, E. halacsyi, one of the rarest and most insular epipactids of the orchid flora of intact, biologically rich natural or seminatural forests of continental Europe, in particular, of the Mediterranean basin, its mountains and approaches.
Key-words: Orchidaceae, genus Epipactis, Epipactis pollinensis, Epipactis viridiflora, Epipactis purpurata, Epipactis pseudopurpurata, Italy, Monte Pollino, Lucanian Apennines, Basilicata, Calabria, Hungary, Bakony Hills, oromediterranean beech forests, biogeography.