Analyses of heliozoan interrelationships: an example of the potentials and limitations of ultrastructural approaches to the study of protistan phylogeny

Abstract

Concepts about specific relationships among various groups of protists are diverse and protist taxonomy is consequently unstable. In order to investigate the causes of this variety, data (mainly ultrastructural) relating to 198 characters of 25 species of heliozoa and other protozoa are analysed. Procedures are used which are compatible with numerical taxonomy (single linkage, complete linkage and group average cluster analyses), cladistic procedures (Camin–Sokal, Dollo and Wagner parsimony analyses) and evolutionary taxonomy (an intuitive tree). The results are presented as branching diagrams. There is no complete congruence among any of the techniques, but all give similar results in some important aspects. The Wagner and Dollo parsimony analyses give those results which are most credible. The results corroborate the view that several major traditional taxa of protozoa (the heliozoa, flagellates, amoebae and filose amoebae) are polyphyletic and require revision. All of the analyses identify the following clusters: actinophryid heliozoa, centrohelid heliozoa, chrysophyte flagellates, actinomonad and pedinellid flagellates and nucleariid filose amoebae. As there is no disagreement, these are confirmed as monophyletic taxa. There is a strong suggestion for a close relationship between dimorphid flagellates and desmothoracid heliozoa. There is some support for the suggestion that the actinophryid heliozoa are more closely related to actinomonad helioflagellates than to other heliozoa. The results are summarized as an unrooted ‘true tree'. The lack of agreement among the analyses appears not to be due to a lack of rigour in analytical procedures, but to an inadequate supply of data. The paucity of data cannot be compensated for by the application of repeatable techniques. Most relationships among high level protist taxa are likely to be (currently) obscured by similar limitations. Ultrastructural data are well suited to mapping out the diversity of protozoa. Electron microscopy currently appears to be the most valuable technique for investigating problems of evolutionary relationships of protists. Various hurdles to the development of a natural (phylogenetic) classification of protists are discussed.