We present a first treatment of the Tortella tortuosa complex for Europe. We analysed molecular relationships based on the nuclear ITS and the plastid atpB-rbcL and rps4 in a network context and thereafter characterized the identified entities by their morphology. We found eight morphologically and molecularly distinct entities at the species level, which are also supported in ASAP analyses of the molecular data; one species includes two varieties. In some cases, nuclear and plastid data suggest different relationships and we found a few likely recent hybrid collections. To the main characters of taxonomic importance belong stem anatomy, leaf shape and papillosity. We describe three species as new: T. commutata (a widespread plant; including the new var. valida), T. dolomitica (known only from the Alps) and T. splendida (an Arctic-alpine element), replacing T. arctica auct. For T. angustifolia and T. robusta (both montane) new combinations at the species level are provided. Tortella bambergeri (a submediterranean element), T. fleischeri (an Alpine element, recurring in Scotland) and T. tortuosa s. str. (widespread) complete this informal group of morphologically similar and partly related species. The species differ in ecological requirements and distribution areas, although mixed stands of two or three species are frequent. The area richest in species in Europe is the Alps with all eight species, whereas we found only four from Scandinavia.

In total, 99 Tortella tortuosa complex specimens were included in the molecular portion of this study, including 72 specimens that were included in earlier Tortella studies by the authors, five specimens which sequences were downloaded from GenBank and 22 specimens which sequences were newly generated for the present study. We studied the nuclear internal transcribed spacers 1 and 2 (ITS) and the plastid atpB-rbcL spacer (atpB-rbcL) and the rps4 gene + trnS-rps4 spacer (rps4). The new sequences were generated as described by Hedenäs (2015: Tortella rigens (Bryophyta, Pottiaceae): relationships, regional variation, and conservation aspects. – Pl. Syst. Evol. 301: 1361-1375.). In the final data set, all three sequences were available for all species except T. alpicola and T. spitsbergensis, for which rps4 was missing.

Nucleotide sequence fragments were edited and assembled for each DNA region using PhyDE® 0.9971 (http://www.phyde.de/index.html; accessed 22 November 2018). The assembled sequences were aligned manually in PhyDE®. Regions of partially incomplete data in the beginning and end of the sequences were identified and were excluded from subsequent analyses. Gaps were coded using the simple indel coding of Simmons and Ochoterena (2000) in SeqState (Müller 2005).

The data includes two files:

tort_final_ITS_sic.fas: ITS alignment with base positions 1-781 and indel coding at positions 783-822.

tort_final_CHL_sic.fas: (A) atpB-rbcL alignment with base positions 1-565 and indel coding at positions 567-581; (B) rps4 alignment with base positions 583-1184 and indel coding at position 1186. 

The files are in fasta format and can be opened by most softwares used in molecular analyses.