Do ploidy level, morphology, habitat and genetic relationships in Alpine Vaccinium uliginosum allow for the discrimination of two entities?

Evolutionary processes such as recurrent origin of polyploids and enhanced gene flow among polyploids make polyploid complexes taxonomically highly intricate. One such complex is the mostly diploid and tetraploid cloudberries (Vaccinium uliginosum L. s.l.), which are not only one of the most frequently recorded dwarf shrubs in the Arctic, but also in mountain ranges such as the European Alps. Surprisingly, in spite of its ubiquity there is very little information on variation in ploidy level, habitat and morphology of V. uliginosum s.l. across the Alps and in adjacent areas; also the taxonomic status of V. uliginosum s. str. and V. gaultherioides is controversial. Based on five latitudinal transects sampled across the middle Alpine arc and an integrative approach combining flow cytometry, plastid DNA sequencing, amplified fragment length polymorphism fingerprinting, morphometrics and characterization of ecological niches via phytosociology and Landolt indicator values the following questions were addressed: (i) What is the large-scale distribution of diploid and tetraploid V. uliginosum s.l. in the area sampled? (ii) Are molecular genetic relationships governed by ploidy level differentiation or is there evidence for acrosscytotype gene flow causing geography-correlated clustering? Is there evidence for (iii) ecological or (iv) morphological separation between cytotypes? We revealed that tetraploids occur throughout the area investigated in the Alps and in their northern forelands, whereas diploids appear to be restricted to the Central Alps and the Southern Limestone Alps. The two cytotypes corresponded to two non-overlapping groups in the AFLP data set, and were also in almost perfect congruence with two plastid lineages. The single evidence for gene flow between diploids and tetraploids was a tetraploid individual carrying a haplotype from an otherwise exclusively diploid lineage. Diploids and tetraploids show some degree of niche separation, which is mostly due to the exclusive occurrence of tetraploids at low altitudes; above 1500 m a.s.l. both cytotypes co-occur. Accordingly, tetraploids occur in a broader range of plant communities than diploids. In spite of the clear separation of diploids and tetraploids in the molecular data sets we failed to detect consistent morphological differences. In particular, absolute size characters exhibited a strong inverse correlation with altitude within the tetraploid cytotype, which prevents their use for reliable determination. Consequently, based on the characters employed here, it is currently impossible to morphologically discriminate diploid V. gaultherioides and tetraploid V. uliginosum s. str. in areas where their ranges overlap, i.e. in subalpine and alpine dwarf shrub communities.