Taxonomy offers precise species identification and delimitation and thus provides basic information for biological research, e.g. through assessment of species richness. The importance of molecular taxonomy, i.e., the identification and delimitation of taxa based on molecular markers, has increased in the past decade. Recently developed exploratory tools now allow estimating species-level diversity in multi-locus molecular datasets. Here we use molecular species delimitation tools that either quantify differences in intra- and interspecific variability of loci, or divergence times within and between species, or perform coalescent species tree inference to estimate species-level entities in molecular genetic datasets. We benchmark results from these methods against 14 morphologically readily differentiable species of a well-defined subgroup of the diverse Drusinae subfamily (Trichoptera, Limnephilidae). Using a 3798 bp (6 loci) molecular data set we aim to corroborate a geographically isolated new species by integrating comparative morphological studies and molecular taxonomy. Our results indicate that only multi-locus species delimitation provides taxonomically relevant information. The data further corroborate the new species Drusus zivici sp. nov. We provide differential diagnostic characters and describe the male, female and larva of this new species and discuss diversity patterns of Drusinae in the Balkans. We further discuss potential and significance of molecular species delimitation. Finally we argue that enhancing collaborative integrative taxonomy will accelerate assessment of global diversity and completion of reference libraries for applied fields, e.g., conservation and biomonitoring.

In this paper, we describe a new species, Potamophylax idliri sp. nov., from the Jastrebac Mountains in Serbia, which is morphologically most similar to Potamophylax coronavirus Ibrahimi, Bilalli & Vitecek 2021 (in Ibrahimi et al. 2021) and Potamophylax juliani Kumanski 1999 (in Kumanski & Malicky 1999). The males of the new species differ from all known species of the Potamophylax winneguthi Species Group by morphological characters of the male genitalia. The genetic distinctiveness calculated based on sequencing of the barcode region of the cytochrome c oxidase subunit I gene (COI) between Potamophylax idliri sp. nov. and other species of the Potamophylax winneguthi Species Group is on par with those of other morphologically recognized species in the group and especially the other four species of the P. winneguthi Species Cluster. Potamophylax winneguthi Species Cluster is erected for the following species: P. idliri sp. nov., P. winneguthi, P. haidukorum Malicky 1999 (in Kumanski & Malicky 1999), P. juliani, and P. coronavirus, characterized primarily by the bilobed apical margins of inferior appendages with pointed corners, in lateral and caudal views. The new species was found at two localities in the Jastrebac Mountains, at 988 and 891 m above sea level. Considering the knowledge about distribution and ecology of the other species of this group, we posit that P. idliri sp. nov. is a microendemic of the Jastrebac Mountains in Serbia.

The presence and distribution of Rhyacophila fasciata Hagen 1859 in Europe were revised, based on bibliographic study, collection specimens, and new material collected in different countries. The status of formerly synonymized species, Rhyacophila ferruginea (Scopoli 1763) and Rhyacophila septentrionis McLachlan 1865 was also assessed. The type of R. ferruginea is missing, the taxon is still unidentified, and thus we propose Rhyacophila ferruginea as a nomen dubium. Morphological features and genetic evidence revealed that R. septentrionis differs from R. fasciata, so we propose to change its status to status resurrectus. We therefore include new descriptions of the different stages (larva, pupa, male, and female) of R. fasciata and of R. septentrionis, together with a molecular analysis based on mitochondrial cytochrome oxidase I (mtCOI) and ecological notes. The species R. coppai Oláh 2020 NEW SYNONYM and R. soreda Coppa & Oláh 2020 NEW SYNONYM are synonyms of R. sociata Navás 1916; the species R. kopasa Oláh & Coppa 2020 NEW SYNONYM and R. rova Oláh & Coppa 2020 NEW SYNONYM are synonyms of R. denticulata McLachlan 1879; the species R. matrensis Oláh & Szczęsny 2020 is probably a synonym of R. fasciata, so more study of this species is needed in order to confirm or deny that it is a valid species.

Deep learning models, known as convolutional neural networks (CNNs), have paved the way for reliable automated image recognition. These models are increasingly being applied in research on freshwater biodiversity, aiming to enhance efficiency and taxonomic resolution in biomonitoring. However, insufficient or imbalanced datasets remain a significant bottleneck for creating high-precision classifiers. The highly imbalanced data, where some species are rare and others are common, are typical of the composition of most benthic communities. In this study, a series of CNN models was built using 33 species of aquatic insects, with datasets ranging from 10 to 80 individuals, to determine the optimal number of individuals each class should have to build a high-precision classifier. We also consider the effect of class imbalance in the training dataset and the use of oversampling technique. The results showed that a robust model with acceptable accuracy (99.45%) was achieved with at least 30 individuals per class. A strongly imbalanced dataset caused an approximately 2% decrease in classification accuracy, while a moderately imbalanced dataset had no significant effect. The application of the oversampling technique enhanced in 1.88% the accuracy of strongly imbalanced models. These findings can help effectively tailor future aquatic macroinvertebrate training datasets.