Compared to Jesenik, Krkonose Mts. are higher (max. altitude: Snezka Mt., 1602), contain larger subalpine areas (about 50 km2) and rocky valleys resembling those of much higher mountains. The main difference in vegetation is a native occurrence of Pinus mugo, a dwarf pine forming a distinct belt sandwiched between spruce forests and subalpine grasslands.
Krkonose lacked truly alpine Erebia until the early 1930s, when the entomologist Josef Soffner sent two parcels, each containing 50 E. epiphron and 50 E. sudetica females, from a Jesenik site to two Krkonose inn-keepers, asking them to release the butterflies in the wild. The target inns were Lucni bouda (1,410 m.) amidst summit grasslands and Rychorska bouda (1,000 m), a rather uncharacteristic site of pastures below the timberline.
Not much was heard about the releases until the 1960s, when Borkowski (1966) reported E. epiphron from Krkonose Northern slopes, about 1 km from the Lucni bouda release site. Soffner (1967) responded by publishing a paper describing the circumstances of the 1930's transfer. Gradually, the butterfly became very abundant within its new range (Liska and Skyva 1997; Cizek et al. 2004). In contrast, neither the released E. sudetica nor the other release of E. epiphron established a population. It is evident from hindsight that Soffner did not understand the precise habitat preferences of the two species, and selected inappropriate settings for both E. sudetica releases and for the Rychorska bouda release of E. epiphron.
For decades, there was no systematic interest in the established E. epiphron population, despite some fascinating questions brought about by its existence. Specifically, the bottleneck during the transfer could have impaired the adaptive ability of butterflies. On the other hand, the release area contains a broader range of biotopes than the native Jesenik area, which might allow expansion to novel habitats. Last but not least, tracking the newly established population could have allowed discerning possible dispersal barriers.
Cizek et al. (2003) approached these questions in a trapping study, similar to the earlier Jesenik study by Benes et al. (2000). Series of traps were set to all main Krkonose biotopes, from clearings deep in the forest belt up to the rocky summits. The mountains were also systematically surveyed to ascertain the current distribution extent. As in Jesenik, E. epiphron abundances peaked at subalpine grasslands near 1,300 m. The species nevertheless reached the uppermost elevations at about 1,600 m and, more surprisingly, has colonized open clearings and hay meadows below the timberline, down to about 1,100 m. Compared to the Jesenik population, the Krkonose population in fact expanded both uphill and downhill.
An important finding concerned the limits of the expansion. Soffner released the butterfly at the Eastern ridge of the Krkonose Mts., which is completely colonized now. The Western ridge, similar to the Eastern ridge in the area of subalpine habitats and maximum elevations, remained uncolonized for over 75 years. The two ridges are separated by about 4 km (aerial distance) of forested elevations, which apparently acted as a dispersal barrier. The butterfly failed to cross the forests -while colonizing the non-forested sites below the timberline.
Another evidence for avoidance of forests originated from a study of the behavior of E. epiphron and E. euryale, conducted in sympatry at the Jesenik main ridge (Konvicka et al. 2002). We recorded the behavior along repeated transects, together with time of day, weather, and vegetation where activities took place. Whereas E. euryale utilized woody structures, such as dwarfed spruce, for activities as resting and basking, E. epiphron restricted all its activities to open grasslands.
Further evidence that forests restrict E. epiphron dispersal originated from a genetic study of the native and introduced populations (Schmitt et al. 2005). The two native populations within Jesenik were significantly differentiated, as were the source population in Jesenik and the introduced population in Krkonose. The genetic distance between the Jesenik source and the Krkonose target sites was less than the distance between samples from northernmost and Southernmost limits of Jesenik main ridge. This revealed isolation by distance along the ridge, consistent with rather limited dispersal. The smaller lateral Jesenik population was strongly genetically impoverished. Its mean heterozygosity was half of the mean heterozygosity of three samples taken along the main ridge (5.5 vs. 10.0%), and similar patterns applied to the total percentages of polymorphic loci, polymorphic loci with the most common allele not exceeding 95%, and the mean number of alleles per locus. No such impoverishment applied to the introduced Krkonose population, indicating that the 50 females transferred by Soffner sufficed to transfer the entire (allozyme) allelic variation of the Jesenik source population.
Perhaps the most intriguing result of the population transfer was the colonization of lower-elevation habitats in Krkonose, as opposed to Jesenik (Cizek et al. 2004). This is explicable by different levels of fragmentation of mountain forests in the two mountains, resulting in different availability of non-wooded habitats within their forested elevations. A peculiar feature of the Krkonose landscape is "cultural"
meadows and pastures, created by forest clearance in altitudes of 800-1,200 m (Krahulec et al. 1997). Moreover, much of Krkonose mountain woodlands fell victim to "salvation logging" of the 1980s, which reacted to die-backs caused by atmospheric pollution (cf. Emmer et al. 1998). The forests are thus dissected by a dense network of open habitats. These are practically missing in the forested elevations of Jesenik, making Jesenik mountain forests more restrictive for dispersing E. epiphron than Krkonose forests (Fig. 2).
The colonization of elevations near 1,100 m in Krkonose, but not in Jesenik, documents that a lower climatic elevation limit of E. epiphron is situated in altitudes suitable for tree growth. The current lower distribution limits are thus not determined solely climatically, but also by the availability of open grasslands below the timberline.
Was this article helpful?