Potentially misleading phylogenetic signal and its explanation in single species studies – contrasting Loeskypnum badium, Sarmentypnum exannulatum and Warnstorfia fluitans

Abstract. Incongruence among morphological and different molecular data sets can be troublesome for species and genus circumscriptions and may cause problems in interpreting intraspecific relationships. Based on the nuclear ITS1 and 2, and the plastid rpl16, trnG and trnL-trnF, it is shown that while such incongruence is not detected if the individual species Loeskypnum badium, Sarmentypnum exannulatum or Warnstorfia fluitans are studied, it may be striking when analyzed in a wider context within the Calliergonaceae. The wider context did not affect the interpretation of internal relationships within L. badium, but many specimens of the other two species were suggested to belong to different species by either marker set. For a large minority of S. exannulatum (8% of specimens), ITS suggested W. fluitans and for 28% of W. fluitans the plastid markers suggested S. exannulatum or (one case) S. sarmentosum. The found pattern agrees with a scenario where archegonia of female plants of S. exannulatum or S. sarmentosum were fertilized by W. fluitans spermatozoids. The produced spores all have the maternally inherited Sarmentypnum plastids, but nuclear genome portions from either the involved Sarmentypnum species or W. fluitans. If foreign plastids affect the functionality of plants less than foreign nuclear portions in plants with mixed genomes, morphological W. fluitans plants with Sarmentypnum plastids should survive to a higher degree than vice versa.