Multiple glacial refugia and postglacial colonization routes inferred for a woodland geophyte,Cyclamen purpurascens: patterns concordant with the Pleistocene history of broadleaved and coniferous tree species

The phylogeographical history of mid‐altitude woodland herbs that depend on moist and shaded forest habitats is poorly understood. Here, we analysed the genetic structure of Cyclamen purpurascens, a mountainous calcicolous perennial, to test hypotheses regarding its glacial survival in single or multiple refugia and postglacial colonization routes, and to explore how they are congruent with the histories inferred for temperate trees and other mountainous herbs. We gathered AFLP data and chloroplast DNA sequences (trnD‐trnT region) from 68 populations spanning the entire distribution range (the Jura Mountains, Alps, western Carpathians, Dinarides). Both genetic markers revealed two main phylogeographical groups (phylogroups) in C. purpurascens. Additionally, AFLP data detected a more detailed structure of five phylogroups: two widespread, showing east?west geographical separation, and three local ones, restricted to somewhat disjunct, marginal regions of the species range. We suggest that C. purpurascens survived the last glaciation in two main regions, the foothills of the Southern Limestone Alps and the Karst area of the north‐western Dinarides, and possibly also in microrefugia in the Western Carpathians. The glacial persistence and colonization routes of this woodland herb are highly concordant with those inferred for several temperate trees, especially the European beech. © 2012 The Linnean Society of London, Biological Journal of the Linnean Society, 2012, 105 , 741–760.     

Genetic and morphological variation in Viola suavis s.l. (Violaceae) in the western Balkan Peninsula: two endemic subspecies revealed

The Balkan Peninsula, with many endemic species, is known as one of the most important speciation and diversification centres in Europe. Here, we present a study of the western Balkan populations of the polymorphic European species, V. suavis s.l., which have been reported under the name V. adriatica, but their taxonomic status and position within the genus have remained uncertain. Viola suavis s.l. and nine close relatives sampled across Europe were subjected to molecular (sequencing of nuclear ribosomal internal transcribed spacers and amplified fragment length polymorphism), karyological and morphometric analyses. Our results revealed the presence of four allopatric, genetically and morphologically differentiated lineages within V. suavis s.l. in Europe, which are suggested here to be recognized at the subspecific rank. Populations from the western Balkans were segregated into two distinct entities: (1) those from north-western Croatia correspond to the previously recognized taxon, V. suavis subsp. adriatica and (2) those from southern Dalmatia (southern Croatia, southern Bosnia and Herzegovina, and south-western Montenegro) are described here as V. suavis subsp. austrodalmatica subsp. nov. The other two lineages of V. suavis s.l., which both harbour blue- and white-flowered morphotypes, occur in central and eastern Europe (V. suavis subsp. suavis) and in north-eastern Spain (plants provisionally treated as V. suavis ‘Spain’). The AFLP and morphological data indicate gene flow between the nominate subspecies and V. suavis subsp. adriatica in a few localities. The distribution of the two western Balkan subspecies is discussed and an identification key to the V. suavis subspecies in Europe is presented.     

Artificial intelligence in pest insect monitoring

Abstract Global problems of hunger and malnutrition induced us to introduce a new tool for semi‐automated pest insect identification and monitoring: an artificial neural network system. Multilayer perceptrons, an artificial intelligence method, seem to be efficient for this purpose. We evaluated 101 European economically important thrips (Thysanoptera) species: extrapolation of the verification test data indicated 95% reliability at least for some taxa analysed. Mainly quantitative morphometric characters, such as head, clavus, wing, ovipositor length and width, formed the input variable computation set in a Trajan neural network simulator. The technique may be combined with digital image analysis.