A bridge too far in naming species: a total evidence approach does not support recognition of four species in Desertifilum (Cyanobacteria).

A population of Desertifilum (Cyanobacteria, Oscillatoriales) from an oligotrophic desertic biotope was isolated and characterized using a polyphasic approach including molecular, morphological, and ecological information. The population was initially assumed to be a new species based on ecological and biogeographic separation from other existing species, however, phylogenetic analyses based on sequences of the 16S rRNA gene and 16S-23S ITS region, placed this strain clearly within the type species, Desertifilum tharense. Comparative analysis of morphology, 16S rRNA gene similarity, 16S-23S ITS secondary structure, and percent dissimilarity of the ITS regions for all characterized strains supports placing the six Desertifilum strains (designated as PD2001/TDC17, UAM-C/S02, CHAB7200, NapGTcm17, IPPAS B-1220, and PMC 872.14) into D. tharense. The recognition of Desertifilum salkalinema and Desertifilum dzianense is not supported, although our analysis does support continued recognition of Desertifilum fontinale. Pragmatic criteria for recognition of closely related species are proposed based on this study and others, and more rigorous review of future taxonomic papers is recommended.