Taxonomic uncertainty and a conundrum: Reassessing the tetraploid taxa of the Dryopteris affinis complex (Dryopteridaceae)

This work is a supplement to a previous publication on the Dryopteris affinis complex in the German Harz Mts. (Bar et al. 2020). It reports on the identification and naming of a tetraploid and population-forming taxon that was called "Dryopteris affinis s. l. (tetraploid, Rehberger Graben)" without a specific name. By extensive investigations of various traits (morphology, nuclear DNA content, size of spores and stomata) we show that Dryopteris cambrensis subsp. pseudocomplexa Fraser-Jenk. (Fraser-Jenkins 2007) is the earliest available name for this taxon. However, D. cambrensis s. str. is triploid, and subsp. pseudocomplexa tetraploid. We, therefore, prefer to use the name available at species level, D. pseudocomplexa (Fraser-Jenk.) P.D. Sell (Sell & Murrell 2018). A total of 15 D. pseudocomplexa plants from different origins (Isle of Skye, the type locality, Isle of Arran, and the Vosges Mts.) were compared with the plants from the Harz Mts. Some of them were examined by flow cytometry already in 2018 and recognized as tetraploid. We also report on mas. Both tetraploids agree in all traits examined. The only exception is the nature of the spores (the hybrid has a higher percentage of aborted spores) and, related to this, the colour of the sporangia at maturity. It appears safe to assume that both taxa have the same genomic composition. A third taxFraser-Jenkins in 2007 and interpreted as the hybrid D. cambrensis subsp. pseudocomplexa x D. filix-mas; the cytotype was declared as unknown. If the assumed ancestry is indeed correct, the hybrid would be hexaploid. The herbarium specimen of the holotype needs to be examined care fully to confirm (or disprove) hexaploidy. Other new findings include the evidence for a strong correlation between ploidy level on the one hand and nuclear DNA content and dimensions of spores and stomata on the other. The nuclear DNA content of the allotetraploid D. filix-mas is lower than expected when assuming the parentage of the diploidD. caucasica andD. oreades. This is most likely due to genome downsizing. Finally , aspects of ecology (plant communities and abi-otic factors) are discussed.