Morphological and phylogenetic data do not support the split of Alexandrium into four genera

A recently published study analyzed the phylogenetic relationship between the genera Centrodinium and Alexandrium, confirming an earlier publication showing the genus Alexandrium as paraphyletic. This most recent manuscript retained the genus Alexandrium, introduced a new genus Episemicolon, resurrected two genera, Gessnerium and Protogonyaulax, and stated that: "The polyphyly [sic] of Alexandrium is solved with the split into four genera". However, these reintroduced taxa were not based on monophyletic groups. Therefore this work, if accepted, would result in replacing a single paraphyletic taxon with several non-monophyletic ones. The morphological data presented for genus characterization also do not convincingly support taxa delimitations. The combination of weak molecular phylogenetics and the lack of diagnostic traits (i.e., autapomorphies) render the applicability of the concept of limited use. The proposal to split the genus Alexandrium on the basis of our current knowledge is rejected herein. The aim here is not to present an alternative analysis and revision, but to maintain Alexandrium. A better constructed and more phylogenetically accurate revision can and should wait until more complete evidence becomes available and there is a strong reason to revise the genus Alexandrium. The reasons are explained in detail by a review of the available molecular and morphological data for species of the genera Alexandrium and Centrodinium. In addition, cyst morphology and chemotaxonomy are discussed, and the need for integrative taxonomy is highlighted.

Pentaplacodinium saltonense gen. et sp. nov. (Dinophyceae) and its relationship to the cyst-defined genus Operculodinium and yessotoxin-producing Protoceratium reticulatum

Strains of a dinoflagellate from the Salton Sea, previously identified as Protoceratium reticulatum and yessotoxin producing, have been reexamined morphologically and genetically and Pentaplacodinium saltonense n. gen. et sp. is erected to accommodate this species. Pentaplacodinium saltonense differs from Protoceratium reticulatum (Claparède et Lachmann 1859) Bütschli 1885 in the number of precingular plates (five vs. six), cingular displacement (two widths vs. one), and distinct cyst morphology. Incubation experiments (excystment and encystment) show that the resting cyst of Pentaplacodinium saltonense is morphologically most similar to the cyst-defined species Operculodinium israelianum (Rossignol, 1962) Wall (1967) and O. psilatum Wall (1967). Collections of comparative material from around the globe (including Protoceratium reticulatum and the genus Ceratocorys) and single cell PCR were used to clarify molecular phylogenies. Variable regions in the LSU (three new sequences), SSU (12 new sequences) and intergenic ITS 1-2 (14 new sequences) were obtained. These show that Pentaplacodinium saltonense and Protoceratium reticulatum form two distinct clades. Pentaplacodinium saltonense forms a monophyletic clade with several unidentified strains from Malaysia. LSU and SSU rDNA sequences of three species of Ceratocorys (C. armata, C. gourreti, C. horrida) from the Mediterranean and several other unidentified strains from Malaysia form a well-supported sister clade. The unique phylogenetic position of an unidentified strain from Hawaii is also documented and requires further examination. In addition, based on the V9 SSU topology (bootstrap values >80%), specimens from Elands Bay (South Africa), originally described as Gonyaulax grindleyi by Reinecke (1967), cluster with Protoceratium reticulatum. The known range of Pentaplacodinium saltonense is tropical to subtropical, and its cyst is recorded as a fossil in upper Cenozoic sediments. Protoceratium reticulatum and Pentaplacodinium saltonense seem to inhabit different niches: motile stages of these dinoflagellates have not been found in the same plankton sample.