More than a decade ago, most of the primary lineages of protistan evolution had been revealed by electron microscopy [45 in 1984 (Corliss, 1984); mostly twigs with few branches], although the branch order was still wide open. As I have noted before (Taylor, 1978, 1994, 1999), the chief problem with the TEM-based tree was that the root could only be hypothesized on the basis of the absence of near-universal eukaryotic characters (mitochondria, ‘9+2’ structures, histones), never a satisfactory approach, since there was no way at the time to distinguish primary absence from loss. Molecular sequence comparisons offered the exciting possibility of rooting the tree with the use of positive genome sequence characters and the search for traces in nuclei of former organelle genomes. This premise has not proved simple to apply.
Unquestionably, it has been the rapid growth of molecular phylogenetics (Taylor, 1994, 1999; Cavalier-Smith, 1995; Patterson, 2000) that has kept the interest in eukaryotic macro-evolution strong and the evolutionary protistan pot simmering: molecular sequencing continues to provide new insights, cement protistan relationships and raise new debates, particularly as different molecules and different methods yield differing results. 5S rDNA trees obviously had anomalous features, attributed to the small size of the molecule and the correspondingly small amount of usable nucleotide sequence variation. LSU rRNA genes are so large and hypervariable in multiple regions that the latter need to be selectively removed in order to study group-level relationships (e.g. Ben Ali et al., 2001). In the 1980s and 1990s, SSU rDNA seemed to be ‘just right’ in size and information content for the latter purpose. In broad features, the trees it generated corresponded well with the main features of the TEM data (Sogin, 1994; Sogin et al., 1996; Taylor, 1994). Now, finally, it seemed that there was a molecule that could be used to resolve which of the ‘absence/loss' choices was actually a primary absence. Some of the amitochondriate groups, such as the diplomonads and parabasalians and microsporidia, were basal in SSU rDNA trees. However, they were long branches and therefore potentially subject to a methodological artefact that made their placement suspect. In any case, new molecular evidence was emerging that revealed the presence of mitochondrial genes in the nuclei of these amitochondriate groups (Keeling, 1998; Roger, 1999).
Once again, in protistan phylogenetic study, a promising new class of data proved not as simple to apply as it first appeared. I am reminded of how, during the TEM revolution, different authorities believed that study of their organelle would be the golden key that unlocked the door to reveal the ‘real tree’ of the eukaryotes (I could certainly now be accused of ‘mitochondriocentricism’, if such a term existed). So too with the genes. As Ben Ali and his colleagues concluded: ‘despite considerable debate on the vices and virtues of various molecular phylogenetic markers, all have their strengths and weaknesses' (Ben Ali et al., 2001, p. 744).
Favourite molecules? Favourite organelles? All this brings to mind the typically memorable remark of New York baseball legend Lawrence ‘Yogi’ Berra: ‘It's déjà vu all over again!’
It is clear that all possible datasets are needed to act as tests of hypotheses generated by each other, and it is unlikely that one set will be equally useful over the whole range of the eukaryotes (Taylor, 1978).
If protistology is now an established field, what has happened to the disciplines that formerly dealt with protists? Of course, they're still there. While demonstrated polyphyly is now deemed sufficient to discard formal systematic groups, there is no compelling reason to eliminate specializations based on them. For example, parasitologists study highly polyphyletic subject taxa and this has not been a problem.
Traditional protozoology, focussed primarily on non-fungal unicellular heterotrophic eukaryotes, but also including ‘phytoflagellate’ groups in which both photosynthetic and non-photosynthetic members exist, is still alive and well. This is evidenced by the highly useful Illustrated Guide to the Protozoa (Lee et al., 2000), covering the traditional groups. Protozoology is now essentially a subset of the groups contained within protistology sensu lato. There are still protozoologists in protistan clothing, even among those using ‘protists' in paper titles but with the contents still traditionally protozoological. In France, protozoologists have long been known as ‘protistologues’, although the scope of their studies has been explicitly protozoological, even when published in Protistologica (now the European Journal of Protistology and truly protistological). The Précis de Protistologie (de Puytorac et al., 1987) is purely protozoological in scope. Despite its title, the Archiv für Protistenkunde (now Protist) has also been protozoological throughout most of its long history. Traditional background-training limitations are understandably difficult to transcend, with occasional lapses even by the enlightened. Alas, the choice of six pre-eminent, contemporary, taxonomically adept ‘protistological ecologists' (Corliss, 1992) was unintentionally restricted to some who have focused on heterotrophic eukaryotes. So too, the citation of 24 publications on ‘protistan ecology’ (Corliss, 2002) only listed those dealing with non-photosynthetic protists, except for one partially so (chrysomonads). Diatoms, dinoflagellates, cryptomonads, silicoflagellates et al. are just as protistan as ciliates or foraminiferans and are ecologically as important, or much more so in aquatic environments.
A wider protistan scope is essential in the study of biodiversity and biocomplexity, as well as ecology. For example, recent claims that protists (‘microbial eukaryotes’) are essentially cosmopolitan, lacking in distinctive biogeographical distributions [e.g. Patterson & Lee (2000) (explicitly for flagellates) and Finlay (2002)], apparently result from the use of examples from extreme environments and the neglect of marine microplanktonic communities that cover 70 % of the Earth's surface. Near ubiquitousness and random dispersal may be semi-true for terrestrial and other harsh-environment protists, but is contrary to a huge body of marine planktonic and sedimentary record literature dating back to the 19th century (see Taylor, 1980b; summed up for dinoflagellates by Taylor, 1987b). Although marine protists are mostly circum-global (‘latitudinally cosmopolitan’; Taylor, 1987b), their dispersal is controlled largely by currents and is strongly influenced by benthic or planktonic habitats, temperature and proximity to the continental shelf (oceanic, neritic). Oceanic, tropical species, for example, are greatly different from polar ice species or even coastal tropical species. Whole groups, such as the acantharians or polycystine radiolarians, are restricted to, or are found predominantly in, subtropical–tropical marine waters. Symbioses of protists, usually photosynthetic cytobionts with non-photosynthetic protist or invertebrate hosts, are far more common in the tropics than in colder waters (Taylor, 1982, 1990). Endemicism is rare, but apparently does exist in South-East Asian waters. Green Noctiluca, containing a symbiotic prasinomonad, are restricted to these high-diversity waters (Taylor, 1990), and so are a few other dinoflagellates, such as Dinophysis miles var. schroteri (Taylor, 1987b).
Contemporary mycologists appropriately focus on the organisms that make up the monophyletic, platycristate lineage (kingdom) composed of chytrids, the Eufungi and, most recently, microsporidians. The other, unrelated fungal-like groups grouped with them in the past (slime moulds/myxozoans, oomycetes, hyphochytrids) are evidently related to the tubulocristate protists, although they are still studied primarily by mycologists for non-phylogenetic reasons.
Phycology still flourishes as a discipline, although several of its major component groups, particularly the ‘Big Three’ (Reds, Greens, Browns), are not closely related, although reds are closer to greens than browns, and the informal group ‘Algae’ is polyphyletic (Taylor, 1978; Bhattacharya & Medlin, 1998; Graham & Wilcox, 2000; and references therein). The strong acceptance of ‘algal protists' into the phycological fold is illustrated by the inclusion of several papers on dinoflagellates in virtually every recent issue of the Journal of Phycology, and the logo of the Phycological Society of America is a rosette of ceratioid dinoflagellates. Green algae (prasinomonads, chlorophytes, charophytes) are indisputably on the same major lineage as metaphytes (streptophytes), good structural, biochemical and molecular evidence that higher plants evolved from within the greens, and have therefore been placed together in the Plantae (Copeland, 1956) or Viridaeplantae (Cavalier-Smith, 1981). The red algae are more isolated, but are arguably not far from the green algae (Ragan & Gutell, 1995). On the other hand, based on sperm flagellar structure and plastid features, the tubulocristate brown algae can reasonably be viewed as derived from, and classificatorily included in, the stramenopiles/heterokonts, a large clade of multiple names and including both photosynthetic and non-photosynthetic member groups. Linking this assemblage with the alveolates would accord with their common possession of tubular cristae (Taylor, 1994, 1999).
The challenge for the future will be how best to realign these disciplines in a way that makes ‘natural’ classificatory sense. For those who continue to work comfortably within the traditional disciplines, I offer this gratuitous homily: act as a protozoologist, mycologist or phycologist, but please think as a protistologist!
Alexander the Great was required to untie the Gordian knot on a chariot harness in order to become the ruler of Asia Minor. According to popular legend, he succeeded by chopping it in two. As a young person, I did not find this particularly edifying, and still do not. It was cheating, and it seems particularly apt that it was a protist, Plasmodium, that terminated not only his conquests, but also him! The eukaryotic evolutionary knot needs to be properly untied. ISEP has made a good start, and I hope that it will be busily unravelling it for a long time to come.
This paper is also published to mark the 80th birthday of our colleague, John O. Corliss, in February 2002. Since John is one of the most active writers on the topics covered here, and since our interests are very similar, although our backgrounds and viewpoints differ somewhat, this tribute seems to be appropriate.