sign in sign up
<<
Home , Allogromia, Ancyromonas, Apusomonas, Arthonia, Bigyra, Ellobiopsis

Hi ul. R iv. ' 1998,', 73, p p . 203-266 printed in the United © Cambridge Philosophical Society 203

A revised six-kingdom system of

T. CAVALIER-SMITH

Evolutionary Programme, Canadian Institute for Advanced Research , Department o f , University o f British Columbia, Vancouver, BC, Canada V6T If4

(.Received 27 M arch 1 9 9 6 ; revised 15 December 1 9 9 7 ; accepted 18 December 1997)

ABSTRACT

A revised six-kingdom system of life is presented, down to the level of infraphylum. As in my 1983 system are treated as a single kingdom, and are divided into only five kingdoms: , Animalia, Fungi, Plantae and . Interm ediate high level categories (superkingdom, subkingdom, branch, infrakingdom, supcrphylum, and infraphylum) arc extensively used lo avoid splitting into an excessive num ber of kingdoms and phyla (60 only being recognized). The two ‘zoological ’ kingdoms. Protozoa and Animalia, are subject to the International Code of Zoological , the kingdom Bacteria to the International Code of Bacteriological Nomenclature, and the three ‘botanical’ kingdoms (Plantae, Fungi, Chromista) lo the International Code of Botanical Nomenclature, Circumscrip­ tions of the kingdoms Bacteria and Plantae remain unchanged since Cavalicr-Smith (1981). The kingdom Fungi is expanded by adding M icrosporidia, because of sequence evidence that these amitochondrial intracellular parasites are related to conventional Fungi, not Protozoa. Fungi arc subdivided into four phyla and 20 classes; fungal classification at the rank of subclass and above is comprehensively revised. The kingdoms Protozoa and Animalia are modified in the light of molecular phylogenetic evidence that Myxozoa arc actually Animalia, not Protozoa, and that mesozoans arc related lo bilaterian . Animalia are divided into four subkingdoms: (phyla Porifera, , , ), Myxozoa, and Bilatcria (bilateral animals: all other phyla). Several new higher level groupings are made in the kingdom including three new phyla: Acanthognatha (, acanthoccphalans. gastrotrichs, ), ( and ) and Lobopoda (onychophorans and tardi- grades), so only 23 animal phyla are recognized. Archczoa, here restricted to the phyla M etam onada and , are treated as a subkingdom within Protozoa, as in my 1983 six-kingdom system, not as a separate kingdom. The recently revised Rhizopoda is modified further by adding more and removing some ‘rhizopods’ and is therefore renamed . The num ber of protozoan phyla is reduced by grouping and (both now infraphyla) as a new subphylum within the phylum alongside the subphylum , which now includes both the traditional aerobic lobosean amoebae and . Haplosporidia and the (formerly microsporidian) melchnikovcllids are now both placed within the phylum Sporozoa. These changes make a total of only 13 currently recognized protozoan phyla, which arc grouped into two subkingdoms : and Neozoa ; the latior is modified in circumscription by adding the , a new infrakingdom comprising the phyla and ). These changes are discussed in relation to the principles of megasvstematics, here defined as that concentrates on the higher levels of classes, phyla, and kingdoms. These principles also make it desirable to rank Archacbactcria as an infrakingdom of the kingdom Bacteria, not as a separate kingdom. Archaebacteria arc grouped with the infrakingdom Posibactcria to a new subkingdom, Unibactcria, comprising all bacteria bounded by a single membrane. The bacterial subkingdom Negibacteria, with separate cytoplasmic and outer membranes, is subdivided into two infrakingdoms: Llpobacleria, which lack lipopolysaecharide and have only phospholipids in the outer membrane, and Glycobac.tcria, with lipopolysaccharides in the outer leaflet of the outer membrane and phospholipids in Its inner leaflet. This primary grouping of the 10 into subkingdoms is based on the number of ccll-cnvclope membranes, whilst their subdivision into infrakingdoms emphasises their mem brane chemistry; definition of the ncgibacterial phyla, five at least partly photosynthetic, relies chiefly on pholosynthelic mechanism and ccll-cnvelope structure and chemistry corroborated by ribosomal RNA phylogeny. The kingdoms Protozoa and Chromista are slightly changed in circumscription by transferring subphylum Opalinata (classes Opalinea, Proteromonadea, Blastocystea cl. nov.) from Protozoa into infrakingdom Heterokonta of the 204 T. Cavalier-Smith

kingdom Chromista. Opalinata arc grouped with the subphylum and the zoofiagcllate Developayella elegans (in a new subphylum Bigvromonada) to form a new botanical phylum () of hctcrotrophs with a double ciliary transitional helix, making it necessary to abandon the phylum name

phylogenetic evidence.

Key winds : M cgasystematics, Bacteria, Protozoa, Anthozoa, Fungi, Animalia, Plantae, Chromista, Eumycota, M y c e to z o a .

CONTENTS

I. Introduction ...... 205 II. Philosophic preliminaries...... 2 1 0 11 ') Darwinian evolutionary classification contrasted with H cnnigian cladification ...... 2 1 0 (2) Naming taxa and ...... 213 (3) Principles of ranking taxa...... 214 (4) The need to weigh and integrate phylogenetic evidence from diverse sources...... 215 III. In defence of Bacteria: the sole primary kingdom of life...... 215 (1) The concept of a bacterium (synonym : )...... 215 (2) Changing views on Archaebacteria ...... 216 (3) The importance of cell structure in bacterial classification ...... 217 (4) Characters im portant in the high-level classification of Bacteria...... 220 (5) Bacterial subkingdoms and infrakingdoms ...... 222 IV. Protozoa, the basal eukaryotic kingdom ...... 226 ¡1) Status of Archezoa, early diverging amitochondrial eukaryotes ...... 2 2 6 (a) Changes in circumscription of the Archezoa ...... 226 (2) The protozoan subkingdoms : Archezoa and Neozoa ...... 229 (3) Demarcation between the two zoological kingdoms: Protozoa and Anim alia ...... 231 (4) Classification of the Neozoa ...... 234 (a) The infrakingdom Sarcomastigota ...... 234 (b) The infrakingdom Alveolata ...... 238 (r) The infrakingdom Adinopoda ...... 238 V. The kingdom Animalia and its 23 phyla...... 239 (1) Radiata, the ancestral animal subkingdom ...... 239 (2) The derived subkingdom Mesozoa ...... 2 3 9 (3) The number of animal phyla ...... 239 (4) A broadened phylum Annelida ...... 240 (5) The pseudococlomatc phyla Nemathelminthcs and Acanthognatha phvl. nov...... 2 4 0 (6) The new phyla Brachiozoa and Lobopoda...... 241 (7) Phylogenetic assumptions behind the new bilaterian infrakingdoms and superphyla 241 (8) N e w a n im a l s u b p h v la a n d in f r a p h y la ...... 243 VI. The kingdom Fungi and its four phyla ...... ‘244 (1) C ir c u m s c rip tio n o f th e F u n g i ...... 244 (2) The trichomycctc origin of ...... 245 (3) Revision of higher fungal classification ...... 245 VII. The kingdom Plantae and its five phyla...... 249 (1) New red algal subphyla ...... 249 (2) New chlorophytc infraphyla...... 251 V III. Chromista, the third botanical kingdom, and its five phyla ...... 251 (1) M ultiple losses of chloroplasts by chrom ists...... 252 ¡2) Transfer of Opalinata to Chromista: the new hclcrokont phylum Bigyra...... 253 (3) The origin of ciliary rclroncmcs and the of the ancestral chrom ist ...... 2 5 4 (4) M itochondrial and ccll-surface evolution in chromists ...... 255 (5) Retroneme loss in Opalinata...... 256 (6) Are there other unidentified chromists lurking within the kingdom Protozoal1 ...... 257 A revised six-kingdom system o f life 205

I X . E n v o i ...... 258 X. Conclusions ...... 258 XI. Acknowledgements ...... 259 X II. References ...... 259

I. INTRODUCTION the kingdom). That paper also made major changes in the circumscription of the existing The idea that nature can be divided into three botanical kingdoms Fungi and Plantae. The six- kingdoms, , vegetable and animal (Lemery, kingdom system of Cavalier-Smith ( 1981 ö ) was later 1675), was popularized by Linnaeus in the eight­ slightly modified (Cavalier-Smith, 1983a) by using eenth century. Although separate kingdoms for fungi the older and more widely familiar name Protozoa, (Necker, 1783), Protozoa (Owen, 1858) or bacteria rather than Protista, for the basal eukaryotic (Enderlein, 1925) were proposed later, the sev­ kingdom [in effect, broadening the kingdom Proto­ enteenth century conception ofjust two kingdoms of zoa of Cavalier-Smith (1981a)] and creating a new life dominated biology for three centuries. The protozoan subkingdom, Archezoa, for putatively discovery of protozoa (Leeuwenhoek, 1675) and primitively amitochondrial protozoa. Since then the bacteria (Leeuwenhoek, 1683) eventually under­ boundaries of the kingdom Protozoa have not been mined the two kingdom system. But general agree­ totally stable: the tax on Archezoa was later ment that the living world must be classified into at (Cavalier-Smith, 1987a) modified by exclusion of least five kingdoms (Margulis & Schwartz, 1988; the Parabasala (irichomonads and hypermastigote Cavalier-Smith, 1989a; Mayr, 1990) was reached flagellates), and the revised Archezoa were segre­ only after the dram atic discoveries made by electron gated from Protozoa as a separate kingdom. At the in the second half of the twentieth same time the bacterial or prokaryotic taxa Eu- century. These conclusively confirmed the funda­ bacteria and Archaebacteria, treated as sub­ mental differences between bacteria and eukaryotes, kingdoms by Cavalier-Smith (1981a, 1983a), w'ere and revealed the tremendous ultrastructural di­ raised in rank to kingdoms, thus creating an eight versity of . Acceptance of the necessity for kingdom system (Cavalier-Smith, 1987a, 1989a, b), several kingdoms also owes much to the systematic which I subsequently advocated (Gavaiier-Smith, synthesis of Copeland (1956), and the influential 1991 a -c , 1993a, 1995a) and which has been adopted writings of Stanier (1961 : Stanier & van Niei, 1962) in certain general works (Gould & Keeton, 1996; and W hittaker (1969). Maynard Smith & Szathmary, 1995). Whether more kingdoms than five are needed The central purpose of the present review is to (Lcedale, 1974), and if so how many, has been reconsider the merits of the earlier six-kingdom debated for over 20 years (Cavalier-Smith, 1978, system in comparison with the eight-kingdom sys­ 1981 a; Möhn, 1984; Corliss, 1994). Even more tem. After discussing the advantages and disadvan­ im portant than the sheer number of kingdoms is the tages of each, it will be concluded that the simpler fact that the definition and circumscription of the six-kingdom classification is preferable as a general recognized kingdoms are not yet agreed by different reference system. Finer phylogenetic discrimination systematists. Phylogenetic advances have been so can be achieved by making better use of inter­ great over the past 30 years, however, that we can mediate-level categories such as subkingdoms and now define monophyietic kingdoms, the circum­ infrakingdoms, w'ithout multiplying the number of scription of which ought to be widely acceptable and kingdoms. Some changes in circumscription of four therefore lead to a much desired stability in the of the six kingdoms are also needed; in the light of macrosystem of life. recent molecular sequence evidence it is necessary to Some years ago, I argued that the minimum remove four groups from the Protozoa and to place number of kingdoms suitable for general purposes them in higher kingdoms; thus Microsporidia are was six (Cavalier-Smith, 1981 a), and that in transferred to the Fungi, Myxozoa and Mesozoa to addition to the previously accepted five kingdoms it the Animalia, and Opalinata to the Chromista. This was necessary to recognize a third botanical king­ makes the kingdom Protozoa substantially more dom, Chromista (mainly comprising the homogeneous. The postulated phylogenetic relation­ and those numerous that, remarkably, have ships between the kingdoms, subkingdoms, and the their chloroplasts located within the lumen of the most deeply divergent infrakingdoms of the present rough , not in the as in system are shown in Fig. 1. 206 T. Cavalier-Smith

ANIMALIA FUNGI CHROMISTA PLANTAE M esozoa Chromobiota Bilata ria A M yxozoa

R adiata E om ycota

■►Alveolata A d in o p o d a Sarcomastigota^«^ I chloroplast h tubular mitochondrial cristae spllceosomal introns; bikonty neoka ryotes

peroxisomesj? A rchezoa PROTOZOA

mitochondria endomembrane system, cytoskeleton tetrakont centrioles, nucleus, mitosis, sex

Em pire PROKARYOTA loss of murein isoprenoid Archaebacteria ether lipids Posibacteria

JJos^^utemnembran^

Glyco bacteria

I lipopolysaccharide BACTERIA

Lipo bacteria

acyl ester membrane lipids murein walls; outer membrane

Fig. 1. Postulated phylogenetic relationships between the six kingdoms (upper case) and their subkingdoms. InCra- kingdoms arc also shown for the two basal paraphylelic kingdoms, as arc some key innovations (within heavy boxes). The four m ajor symbiogenctic events in the are shown with dashed arrows: (1 ) the svmbiogenetic origin of mitochondria from an «-proteobacicrium, (2) the probable origin of peroxisomes from a posibacterium (Cavalier- Smith, 1990), (3) the monophyletic origin of chloroplasts from a cyanobacterium, and (4) the origin of the ancestral chromist as a -eukaryotc chimaera between a rhodellophte red algal and a bidliated prolisl host; whether the host was a protozoan, as assumed here, or a very early plant is unclear For clarity infrakingdoms are not shown for the four higher holophvletic kingdoms, Animalia, Fungi, Plantae, and Chromista. The diagnoses of the taxa are given in Table 1 and their composition is summarized in Tables 2-7. Chloroplasts arc assumed to have originated in the latest common ancestor of Plantae, Discicristata and the dinoflagellata protozoa (Cavalier- Smith, 1982), and the of euglcnoids and alvcolates are assumed lo have arisen divergently from this primary endosymbiosis. However, the alternative possibility that and/or euglenoids obtained their chloroplasts secondarily by lateral transfer from a chromobiote and green alga respectively, though considered distinctly less likely, cannot currently be excluded (Cavalier-Smith, 1995 a) ; the source of the non-pholosvnthetic of coccidiomorph Sporozoa is equally uncertain (M cFaddcn & W aller, 1997). For simplicity a fifth important case of svmbiogcnesis is not shown: secondary symbiogenctic lateral transfers of green algal, possibly ulvophytc (Ishidael al., 1998). chloro­ plasts into a ccrcozoan (sarcomastigota host to create the algal Chlorarachnea (e.g. Chlorarachnion). The least certain features of the tree arc the position of its root (some authors favour a root within the Unibacteria rather than the Negibacteria) and of the Discicristata (18S rRNA and EF I-a ti ces put them below the Sarcomastigota, whereas other place them as shown). A revised six-kingdom system o f life 207

When establishing the kingdom Chromista contrast, should be included in the kingdom Fungi. (Cavalier-Smith, 1981a) I mentioned that, as dis­ This circumscription of the kingdom Fungi (see also cussed earlier by Taylor (1978) and Hibberd (1979), Cavalier-Smith, 19876) is now almost universally there were reasons for thinking that proteromonads accepted by phylogenetically oriented mycologists may be closer to Chromobiota than to Protozoa. (Bruns, W hite & Taylor, 1991), and adopted by the They had some ultrastructural similarities with authoritative Dictionary of the Fungi (Hawksworth et chromists and some with the opalinids, so it appeared a l., 1995). It will, therefore, come as a considerable that they might be phenotypically intermediate surprise to mycologists to learn that the kingdom between chromists and protozoa. In to make Fungi now needs to be drastically expanded by the the definition of the Chromista as sharp as possible, addition of the Microsporidia, a phylum of minute I conservatively kept the proteromonads in the intracellular parasites of animals. Though Micro­ kingdom Protozoa, placing them together with sporidia have chitinous spores and are not phago- opalinids and cyathobodonids in the protozoan trophic, they were traditionally thought of as group Proterozoa, but recognized that the boundary protozoa rather than fungi because their vegetative between chromists and protozoans might require cells lack cell walls. Like Opalinata, the Micro­ slight adjustment in future. Patterson (1985) subse­ sporidia lack peroxisomes; but as they also lack quently placed proteromonads and opalinids alone mitochondria they were postulated to be archezoan together in the order . Later Patterson Protozoa (Cavalier-Smith, 1983a); how'ever, recent (1989) argued that slopalinids should be grouped protein sequence data (Li et al., 1961; Edlind et al., with as . 1 have since 1996; Keeling & Doolittle, 1996; Roger, 1996; argued that the Proteomonas is less closely G e rm o t et al, 1997) strongly suggest that they are related to Opalinida than is the genus actually highly degenerate fungi, which have sec­ and placed the orders Karotomorphida and Opal­ ondarily lost mitochondria, like the rumen fungi inida together in the class Opalinea (Cavalier- (Neocallimastigales), which have also been mis­ Smith, 1993a, b) but put proteromonads in a takenly classified as protozoa in the past. I here separate class. As part of a reappraisal of the remove the Microsporidia from the Archezoa and Proterozoa stimulated by recent rRNA (ribosomal place them instead in the kingdom Fungi, where I RNA) sequences (Cavalier-Smith & Chao, 1995, group them with the Archemycota in a new 1997) I have created a new class Proteromonadea subkingdom Eomycota. As the higher level classi­ for the Proteromonadida, abandoned the taxon fication of the kingdom Fungi needs some revision, I Proterozoa and grouped Opalinea with Protero­ divide it into two new subkingdoms (Eomycota and monadea in the subphylum Opalinata (Cavalier- Neomycota) and create a few other new high level Smith, 1997a). In order to increase the homogeneity fungal taxa, and aiso validate some of those proposed of the new protozoan subkingdom Neozoa and the earlier (Cavalier-Smith, 19876). I present a detailed zooflagellate phylum Neomonada, I excluded the classification of the kingdom Fungi, down to the revised Opalinata from them as a taxon level o f subclass. incertae sedii. Here, I accept the view (Patterson, The thiTd refinement made here to the circum­ 1989) that Opalinata (a subphylum now compo- scription of the Protozoa compared with the earlier sitionally the same as the order Slopalinida) are six-kingdom system (Cavalier-Smith, 1983a) is the probably secondarily modified heterokonts, and transfer of the Mesozoa and Myxosporidia into the therefore place them within the chromist infra- kingdom Animalia, as briefly indicated earlier kingdom Heterokonta and group them with the (Cavalier-Smith, 19956, Cavalier-Smith et al., heterotrophic subphylum Pseudofungi (Cavalier- 1996 a). Smith, 1986a, 19896) to form a new heterotrophic The reasons for the above changes in circum­ chromist phylum, Bigyra, in which I also place the scription of the kingdoms Protozoa, Chromista, recently discovered distinctive protist Fungi and Animalia will be explained in detail. In Developayella elegans (Tong, 1995), for which I create view of the numerous changes in high-level classi­ a new order, class and subphylum. fication over the past 15 years, I shall present a In the system of Cavalier-Smith (1981 a) I refined detailed summary of this revised six-kingdom classi­ the kingdom Fungi by removing oomycetes, hypho- fication of the living world, dowm to the level of chytrids, and thraustochytrids and transferring them infraphylum. into the Heterokonta within the new kingdom My redefinition of the kingdom Plantae to include Chromista, and arguing that , by all Viridaeplantae (green ), Glaucophyta, and 208 T. Cavalier-Smith

Table 1. The revised six-kingdom system o f life

Empire or Superkingdom 1. PROKARYOTA**Dougheriy 1957 slat. nov. Allsopp 1969 ( in the same com partm ent as the usually circular chromosome). Kingdom Bacteria** Cohn 1870 stat. nov. Cavalier-Smith 1983 (syn. Procaryotae M urray 1968: no internal cytoskeleton, endom em branc system, nuclei, mitosis or true sex). Subkingdom 1. Negibactcria* Cavalier-Smith 1987 (outer membrane present; acyl ester lipids; with small signal- récognition particle RNA). Infrakingdom Lipobactcria* new infrakingdom (diagnosis: no lipopolysaccharide in outer membrane; murcin cell wall). Infrakingdom 2. Glycobaclcria* new infrakingdom (diagnosis: lipolvsaccharidc in outer membrane; cell wall of murcin or, rarely, protein). Subkingdom 2. Unibacteria* new subkingdom (diagnosis: bacteria with no outer membrane; with large signal- recognition particle RNA as in eukaryotes). Infrakingdom 1. Posibacteria* Cavalier-Smith 1987 stat. nov. (acyl ester lipids; often with tcichoic acids). Infrakingdom 2. Archacbactcria Woesc & Fox 1977 stat. nov. (isoprcnoid ether lipids; murein absent).

Empire or Superkingdom2. E U K A R Y O T A (cytoskeleton, endomembranc system, nucleus, sex). Kingdom 1. Protozoa**G o ld fu ss 1818stat. nov. Owen 1858em. (phagotrophs primitively without plastids, collagen or chitinous vegetative cell walls; unicellular, plasmodial or colonial). Subkingdom 1. Archezoa** Cavalier-Smith 1983 cm. (kinetid letrakont; mitochondria absent; unsplit). Subkingdom 2. Neozoa** Cavalier-Smith 1993 stat. nov. 1997 (kinetid typically ; tubular or rarely flat mitochondrial crisiae; spliccosomal introns common). Infrakingdom 1. Sarcomastigota** Cavalier-Smith 1983 slat. nov. em. (kinetid bikont or secondarily unikont or absent; m itochondria usually with tubular cristae, rarely flat (typically non-discoid) or vesicular; cortical alveoli absent; when present non-erupiivc; axopodial microtubules if present, not spirally or hexagonally a r r a n g e d j. Infrakingdom 2. Discicristata new infrakingdom (kinetid letrakont or bikont; mitochondria or present; cristae discoid; genes mostly unsplit; pscudopodia eruptive when present; cortical alveoli and axopodia a b s e n t). Infrakingdom 3. Alveolata Cavalier-Smith 1991 [cortical alveoli (rarely secondarily absent) ; mitochondrial cristae tubular or ampulliform]. Infrakingdom 4. Actinopoda Calkins 1902 stat. nov. Cavalier-Smith 1996 (axopodia with hexagonal or spirally arranged microtubules; cilia abscnl or for dispersal only; possibly polyphyletic). Kingdom 2. Animalia Linnaeus 1758 em. Cavalier-Smith 1995 (unnecessary synonym M etazoa Haeckel 1874) (ancestrally phagotrophic multicelis with collagenous connective between two dissimilar epithclia). Subkingdom 1. Radiata** Linnaeus 1758 stal. nov. em. Cavalier-Smith 1983 (multicellular animals with radial or biradial symmetry; no anus). Infrakingdom 1. Spongiaria* De Blainville 1816 (choanocytes line body cavity; nervous system primitively absent: ). Infrakingdom 2. Coelenterata* Lcuckarl 1847 cm. aucl. (nerve net). Infrakingdom 3. Placozoa infraking. nov. (without body cavity, gut or nervous system). Subkingdom 2. M yxozoa Grassé 1970 stat. nov. (secondarily unicellular parasites of bilateral animals; spores multicellular; cilia absent: myxosporidia). Subkingdom 3. Hatschck 1888 stat. nov. Cavalier-Smith 1983 (bilateral, primitively with anus). Branch 1. Protostomia* Grobben 1908 (blastopore becomes mouth). Infrakingdom 1. Lophozoa* new' infrakingdom (primitively sessile with U-shaped gui and ciliated oral tentacles with coelomic extensions; early ciliated larvae trochophorcs, later often bivalvcd). Infrakingdom 2. Chaetognathi Leuckart 1854 stal. nov. (non-ciliatcd; thin cuticle not moulted; embryonic cnterocoel absent in adult). Infrakingdom 3. new infrakingdom (non-ciliated; thick cuticle that is moulted; hacmocoel or pseudocoel). Infrakingdom 4. Plalyzoa new infrakingdom (secondarily acoelomate worms without vascular system; ancestrally ciliated). Branch 2. Deuterostomia Grobben 1908 (blastopore becomes anus). Infrakingdom I. Coelomopora stat. nov. (trimeric coelom, anterior compartment with external pore; enterococlous; ciliated pelagic larvae; nerve net). A revised six-kingdom system o f life 209

T a b l e 1- ( co n t.)

Infrakingdom 2. Chordoma Haeckel 1874 em. Hatschck 1888 (sehizococlous or coelom absent; larvae tadpole-like; hollow dorsal nerve cord). Subkingdom 4. Mesozoa Van Beneden 1877 stat. nov. (ciliated multiceilular parasites; no nervous system or gui). Kingdom 3. FungiLinnaeus 1753 slat. nov. Nces 1817 em. (vegetative and/or spore cell walls ofchitin and ß - glucan; no phagocytosis). Subkingdom 1. Eomycota** subking. nov. (diagnosis: hyphac without perforate septa; without dikaryotic phase: ccllulae non binuclcatae; septa usilaic absens; si praesens non perforata). Subkingdom 2. Ncomycola subking. nov. (diagnosis: usually with a dikaryotic phase; hyphae, if present, with perforate septa: cellulac binucleata? plcrumquc pracscntes; endospora aut ascospora aut basidiospora instructa). Kingdom 4. Plantae H a e c k e l 1866em. Cavalier-Smith 1981 (plastids with double envelope in cytosol; starch; no phagocytosis). Subkingdom 1. Biliphyta* Cavalier-Smith 1981 (phyeobilisomes; single thylakoids; starch in cytosol). Infrakingdom 1. Glaucophyta infraking. nov. (diagnosis: pcptidoglycan in plaslid envelope: plastidac pcptidoglycanum instructae). Infrakingdom 2. Rhodophyta infraking. nov. (sine peplidoglycano: plastid envelope lacks pcptidoglycan). Subkingdom 2. Viridaeplantae Cavalier-Smith 1981 (chlorophyll a a n d b ; thylakoids slacked; starch in plastid strom a; ciliary transition nine-fold star: green plants). Infrakingdom 1. Cavalier-Smith 1993 (). Infrakingdom 2. Cormophyta Endlicher 1836 stat. nov. (). K in g d o m5 . C h r o m is t a Cavalier-Smith 1981 em. (chloroplasts with chlorophyllc inside a periplastid membrane within the rough endoplasmic reticulum lumen and/or with rigid bipartite or tripartite ciliary hairs). Subkingdom 1. Cryptista Cavalier-Smith 1989 (ejectisomes; usually with bipartite hairs on both cilia, tiueleomorph and phycohilins; cristae flattened tubules; pellicular plates). Subkingdom 2. Chromobiota Cavalier-Smith 1991 (cristae tubular; posterior often autofluorescent; no nucleomorph, ejectisomes, pellicular plates or phycobilins). Infrakingdom 1. Heterokonta Cavalier-Smith 1986 stat. nov. 1995. em. (rigid tripartite or bipartite hairs on anterior cilium only or, rarely, on cell body; no haplonema). Infrakingdom 2. Haptophyta Cavalier-Smith 1995 (haptonem a; hairs unipartitc tubules, knob hairs or absent).

* Probably paraphyletic taxon. ** Almost certainly paraphyletic. The classification of each of the kingdoms is given in more detail in Tables 2-7, with examples of representatives of the major taxa. It is doubtful if the modification of the spelling of the name Archaebacteria in Bergey ¡ Manual to Archaeobacleria (Gibbons & M urray, 1978) was correct; if archae- was derived from the Greek archaias (ancient) then it is a stem and the ‘o ’ should be inserted according to the rules of the botanical code but this is not required by the bacteriological code; but if it was derived from the Greek arche (beginning), Archaebacteria is a correctly formed pseudocompound word In which archae retains the case ending -ae, a n d as archae is a whole word (not the ste m arch-) it would be incorrect lo insert the ‘o ’. For stability, I prefer lo keep the original spelling and to assume that Woese & Fox (1977a) made no error. The more recent change to (Woese, K andier & VVhcclis, 1990) is both totally unnecessary and highly undesirable; its use should be most strongly discouraged (Cavalier-Smith, 1 9 9 2 6 ).

Rhodophyta, and these three groups alone (Cava- from a single photosynthetic common ancestor, I lier-Sm ith, 1981 a), is not yet w idely accepted, largely think that it will be only a m atter of tim e before the because of a formerly widespread belief in the logic of accepting the of the kingdom polyphyletic symbiotic origins of chloroplasts P l a n t a e sensu Cavalier-Smith (1981a) will also be (Mereschkowsky, 1910; Margulis, 1970, 1981; generally recognized, as it already has been by Raven, 1970). But as the theory of the m onophyletic Ragan & Outtell (1995). I have therefore seen no symbiogenetic origin of chloroplasts (Cavalier- reason to modify the circum scription of the kingdom Smith. 1982), upon which my circumscription of Plantae or its subdivision into two subkingdoms Plantae was based, is increasingly widely accepted (V iridaeplantae and Biliphyta) since Cavalier-Sm ith (Bhattacharya & M edlin, 1995), and as it logically (1981a), and expect the kingdom to continue to be follows that the three m ajor plant groups are derived stable in the future. 21 0 T. Cavalier-Smith

An overview of the revised six-kingdom system is was complimented by certain cladists who mis­ given in Table 1, which includes all thesupraphyletic takenly thought I was applying the principles of taxa, with diagnoses showing how they may be Hennig (1966) because I was using cladistic reason­ distinguished. In oider to save space in the later ing and advocating a phylogenetic classification tables dealing with the phyla, subphyla and infra- with strictly monophyletic taxa. However, I had not phyla of each kingdom diagnoses are given only for then even heard of Hennig or his confusing redefini­ new or substantially emended taxa. Despite the tion of the term monophyletic, which I do not creation here of five new phyla, I accept only 60 accept. Like classical phylogeneticists, I use the word phyla for the entire living world, many fewer than monophyletic to include both holophyletic (Ashlock, the 92 recognized by Margulis and Schwartz (1988) 1971, i.e. monophyletic sensu Hennig, 1966) and or the 100 used by M öhn (1984). I am not actually paraphyletic. I do, however, accept Hennig’s redefin­ a ‘born splitter’ (Corliss, 1994). are best ition ofHuxley's (1957, 1959) term , which was thought of as laterally transmissible parasitic genetic originally defined using the classical, not the elements, not as living organisms, so I do not attem pt Hennigian definition of monophyly. (The earlier use to classify these im portant biological entities here. of the word clade in biology by Lankester (1911 : 1031) as an anglicization for Haeckel’s (1868) taxonomic category of cladus, equivalent in rank to II. PHILOSOPHIC PRELIMINARIES infraphylum in the present system, is no longer current: the Renaissance usage o f‘clade’ to denote (1) Darwinian evolutionary classification a plague or disaster is even more obsolete.) Hennig contrasted with Hennigian cladification and his followers have done much to increase the Ever since Darwin, it has been accepted that ‘ the rigour of phylogenetic analysis, but this advance has systemadst is primarily a morphologist whose task is been brought at the cost of some verbal confusion to discover the true phylogenetic relationships’ as and much harmful dogmatism. Furthermore, though Cummings (1916: 254) noted 80 years ago. Darwin­ cladistic analysis is simply formalized phylogenetic ian evolutionary classifications take into account common sense, and should be strongly encouraged, both the branching patterns and the degree of Hennigian has two grave defects, one change along different branches, and are therefore practical and one theoretical. truly phylogenetic (Mayr & Ashlock, 1991) rather Practically, it leads to serious instabilities in than purely cladistic or purely phenetic. A balanced classification and nomenclature. The great practical classification aims to subdivide a taxon like Eukary- merit of a Darwinian evolutionary classification is ota into a number of subordinate taxa of lower rank that it should be much more stable, since whether a in such a way as to maximize the degree of similarity taxon is holophyletic or paraphyletic is irrelevant to within each taxon, while ensuring that none of them its classificatory role. There are many thousands of is poiyphyletic. Since organisms can vary in many taxa where we do not know whether they are different characters, and in ways varying from the paraphyletic or holophyletic. According to trivial to the profound, it is essential to weigh their Hennigian principles, whenever any taxa become relative importance. Lamarck (1810) initiated the clearly established as paraphyletic we would have to idea of genealogical classification, but it was Darwin split them up, abandon their names and invent two (1859), Haeckel (1866, 1868) and Lankester (1877) or more new taxa. This would be especially who first really popularized the notion of phylo­ nomenclaturally destabilizing for genera, a large genetic classication as the central goal ofsystematics. proportion of which may be paraphyletic. Haeckel (1866) introduced the terms phylogeny, Theoretically, the Hennigian attempt to restrict phylum, monophyletic and cladus, whilst Lankester taxa to clades, and forbid paraphyletic groups is (1977) gave us the concept of a grade and en­ incompatible with the basic purpose of phylogenetic couraged the English-speaking world to substitute classification, even though it misleadingly mas­ Haeckel’s term phylum for subkingdom. These early querades under that name. What a biological pioneers of phylogenetic classification used both classification aims to do is to arrange organisms in a grades and clades as taxa in their influential hierarchical of nested taxa, in which each classifications. more-inclusive higher-level taxon is subdivided com­ When I first argued that the kingdoms of life prehensively into less-inclusive taxa at the next level needed reclassifying on phylogenetic principles to below'. There are two key words here: hierarchical make them monophyletic (Cavalier-Smith, 1978), I and comprehensive. A set of nested clades is A revised six-kingdom system o f life 211

Cladification Classification

C lade E Taxon E C lade D Taxon D Taxon F* C lade A Taxon A C lade B Taxon B Taxon H* Taxon G* C lade C Taxon C

Fig. 2. Contrast between (a) cladification and (b) classification. From the same with holophyletic recent 1 lo 6 and paraphyletic ancestral species 7 to 15 one can produce either a list of nested clades or a comprehensive classification. In the classification taxon D is comprehensively subdivided into three taxa oflow er and equal rank (A, B, F) ; A and B are holophyletic and F is paraphyletic. In the cladification, clade D is subdivided into two clades A and B with the same composition as the holophyletic laxa A and B plus a residue including species 10 lo 12 which is not placed in any subgroup of clade D. The cladification is not a comprehensive classification of clade D, because this paraphyletic residue comprising species 10 to 12 is not listed under clade D even though it is part of that cladc. The size of this paraphyletic residue could be reduced in an alternative cladification by moving the boun­ daries of clades A and B from just below species 7 and 8 to just above species 12; but as species 12 is ancestral to both clados A and B, it cannot be included in either. No clade can be comprehensively subdivided into subdades. A cladification can never be a comprehensive classification since it omits the paraphyletic residues which a compre­ hensive classification must include (taxa F. G and H in this ease). The cladification on the left is congruent with the classification on the right. W hen comprehensively subdivided into taxa of equal rank, it is logically inescapable that every monophyletic taxon (whether holophyletic or paraphyletic) must include at least one paraphyletic taxon. A cladification is a most useful step towards a phylogenetic classification but it does not constitute one in its own right. If paraphyletic taxa are explicitly identified in a phylogenetic classification one can reconstruct the cladification on which it is based by deleting them. But one cannot move from a phylogenetic tree to a classification without exercising taxonomic judgement, which is necessary both in deciding which clados should be treated as taxa and whether the paraphyletic residue should be treated as a single taxon or subdivided. Judgement is likewise needed about the appropriate rank for each taxon. hierarchical, but because it is not comprehensive at level groups. Cladists have long accepted that the each level in the hierarchy, it is much too incomplete inability to classify ancestral and m any fossil taxa is to be properly called a classification. Ernst Mayr the Achilles heel of Hennigian classificatory prin­ ipers. comm.) has recently called such a set of nested ciples, and refer to it as a problem; it is not a clades a cladification. Figure 2 contrasts the way in problem at all for systematics, but merely a basic which a Darwinian evolutionary classification sub­ defect in the Hennigian ideas on classification. divides a phylogenetic tree to produce a com­ Obviously, if you assert that you must not make prehensive classification of the organisms making up paraphyletic groups then you cannot properly the tree, whereas Hennigian cladification of the same classify ancestral species excluded from a particular organisms yields an incomplete subdivision of higher- clade. No comprehensive phylogenetic classification 212 T. Cavalier-Smith

is even theoretically possible unless one accepts as much more artificial than a monophyletic group­ paraphyletic as well holophyletic laxa. The dogma ing. For monophyletic groups (whether paraphyletic against paraphyletic taxa is logically incompatible or holophyletic), their coherence is the result of with the acceptance of both evolution by descent historical genetic continuity between all members of and the goal of taxonomy as the creation of a the group : it is their discontinuity from other groups comprehensive phylogenetic classification of all that is artificial (except for biological species). The organisms, both extant and extinct. claim that a cladogram or phenogram can be As it is theoretically unsound, and in practice converted into a classification without artificial cuts increases nomenclatural and classificatory instabil­ across the natural phylogenetic continuum (Nelson ity, the dogma against paraphyletic groups has been & Platnick, 1981} is at best a spurious half truth; most detrimental to systematics. Cladists, rightly cladograms and phenograms can be claimed to be as immensely impressed with the logic and great value discontinuous as classifications only because the cuts of Hennigian principles for , have across the phyletic continuum have already been naturally been reluctant to accept that they can be made by classical taxonomists when creating the so fatally flawed when applied to taxonomy. Some discrete taxa that form the raw materials for cladistic do recognize this central flaw, but say that in and phenetic analysis. The claim involves an element practice most taxonomists deal only with extant of myopic self-deception that allows some cladists to organisms or that it is never possible to recognize claim falsely that Hennigian systematics is more ancestors with certainty. However, we sometimes natural than classical phyletic taxonomy (Stuessy, can recognize ancestors and need to classify them. 1990). No classification can be totally natural. Furthermore, not all ancestors are extinct. In plants, Whether a taxon is paraphyletic or not is new allopolyploid species are created by hybrid­ irrelevant to its validity as a taxon. It is also ization between two ancestral species followed by irrelevant to m any of the uses to which classifications polyploidization. In several instances, both ancestral are put, such as arranging specimens in a museum, species have been unambiguously identified and organising the chapters in a biology textbook, or both are extant. A well known example is the providing a convenient label, e.g. bacteria or fungi, formation of the new grass Spartina anglica b y for a group of similar organisms. But the distinction allopolyploidy from the ancestorsS. alterniflora a n d S. is im portant in certain uses to which classifications m aritim a. The two ancestors are both perfectly good are sometimes put. Two common examples are the paraphyletic taxa, and as biological species are choice of model systems and the study of group natural taxa, distinct from each other and from their extinction. descendant species S. anglica. Paraphyletic species A wishing to choose a protozoan as a arc therefore as natural as holophyletic ones. Apart model system for understanding animal from the special case of biological species, paraphyl­ would be quite mistaken in supposing that as all etic and holophyletic taxa are both partly artificial protozoa are classified in the same kingdom it does constructs of taxonomists (whether they are asexual not matter which is chosen. Obviously, some species or taxa of higher rank). From such a members of such a paraphyletic group are cladi- perspective a holophyletic taxon is not created by stically more closely related to a derived taxon than the origin of a new synapomorphy, but by the others: for example, would be decision of a systematist to subdivide an historically much more similar to animal cells than would continuous phylogenetic tree at that point. As I have . Likewise, the extinction of all m em ­ stressed previously (Cavalier-Smith, 1993a), every bers of a paraphyletic group, in contrast to that for such cutting off of a holophyletic branch necessarily a holophyletic group, does not constitute extinction simultaneously generates a paraphyletic stem. If I of the entire lineage. But the mistaken assumption cut a tree into pieces, each piece, whether it is a that it does, like the assumption that all members of term inal piece with one cm or a stem piece with two a group are equally related to another group, are or more, is both artificial and natural. Each misuses, not uses, of a classification. W here precise continuous segment owes its continuity to natural phylogenetic relationships are important for a processes, whereas its separation from others is the scientific problem, scientists should base their reason­ result of hum an, i.e. artificial intervention. If I group ing directly on phytogenies and not use classifications together cut pieces that were never joined on the as a crude surrogate for the real thing. The purpose tree, such joining would be totally unnatural, which of classification is to provide a simplified reference is why polyphyletic grouping has long been accepted system that is biologically sound and widely useful. A revised six-kingdom system o f life 213

It should be compatible with the phylogeny, but it advances. This requires some new names of higher cannot serve its central simplifying purpose unless it taxa and some shifts in the meaning of old ones. leaves out some of the fine detail about relationships Because classifications are scientific generalizations that are essential for some phylogenetic purposes. about the varied properties of organisms, and not One can use a phylogeny as a basis for making a just indexes to stored information, they must be classification, but one cannot logically deduce a fully changed if new knowledge shows them to be detailed phylogeny from a classification. Nor is a fundamentally wrong. During the past half century, phylogeny sufficient to give a classification. A electron microscopy, biochemistry and molecular phylogeny and a classification must be congruent sequencing have immensely deepened our under­ (i.e. not contradictory) but they arc different ways of standing of biodiversity. After the revolutionary abstracting from and representing biological relation­ changes occasioned by this have been soundly ships. It is highly desirable that taxonomists publish assimilated, our classifications will become more an explicit phylogeny and statement of their phylo­ stable. Thus stability is a valuable outcome of a genetic assumptions at the same time as their sound classification, but must not dominate system­ classification, a practice I have long followed atics by impeding necessary changes. (Cavalier-Smith, 1978). This will allow users to use It is im portant to have a brief name for all cladcs whichever is most appropriate to their needs. In of major evolutionary significance, but one should order to alert users of classifications to the sometimes not make the mistake of supposing that all such important differences between holophyletic and clades need to be made into taxa. Nor should one paraphyletic taxa it would be good practice if, where make the converse error, especially common among known, these are clearly distinguished. In Tables cladists, of supposing that only clades can be taxa. 1-7 in this review I mark those taxa that are almost Taxa and clades should be thought of as two partly ceTtainly paraphyletic and those which, in my overlapping sets that serve somewhat different, but phylogenetic judgem ent, are probably paraphyletic. equally essential, biological purposes. In a given Taxa not so identified include not only those known classification such as the six-kingdom system advo­ or thought to be holophyletic but also those for cated here, some taxa (e.g. Fungi, Animalia, which evidence is too little for me to make a Chromista and probably Plantae) are clades, but judgem ent; future work is bound to show that some paraphyletic taxa like Bacteria, Archezoa and are paraphyletic, and others polyphyletic, and that Protozoa are not. For example, an im portant clade at least some of those thought to be paraphyletic are which is not a taxon in this system is the clade actually holophyletic. comprising Animalia, Fungi and , which I named opisthokonta (Cavalier-Smith, 1987A) ; molecular phylogeny has since confirmed the validity (2) Naming taxa and clades of this clade (Cavalier-Smith, 1993a; W ainright et A named taxon such as Protozoa or Plantae has to a i , 1993). I deliberately did not create a taxon serve a dual role. Its name acts as a label to refer Opisthokonta or give a diagnosis for it. This is succinctly to a group of related organisms sufficiently because opisthokonta are far too phenotypically similar to each other that they share a generalized diverse to be useful as a major unit of eukaryote phenotype, by which they can be readily recognized classification. One cladist has even told me that I and distinguished from other taxa. Secondly, they ought to have created a formal kingdom Opis- are units that can be grouped together hierarchically thokonta. A kingdom Opisthokonta would be much to create a smaller number of higher taxa. A list of less useful than the existing kingdoms Animalia, nested clades, such as that in Patterson (1994), can Fungi and Protozoa as a way of subdividing the be very useful as a phylogenetic summary, but it is living world into manageable major groups of similar not the same as a classification. organisms, i.e. in classification as opposed to phy­ Stability is desirable in nomenclature, so I have logeny. The cladist’s suggestion reflects a misun­ tried wherever reasonably possible to retain older derstanding of the purposes and functions of classi­ and more familiar names. But stability is not a fication, which is all too common among cladists, prim ary value in classification. If it were, we should who are typically very much more interested in rigidly retain the oldest classification irrespective of phylogeny than in classification, and often forget how bad it is! Especially among that these are two different branches of systematics. traditional classifications have been so grossly in­ Clades and taxa are non-equivalent concepts, as is correct that they must be changed as science shown pictorially in Fig. 3. 214- T. Cavalier-Smith

(3) Principles of ranking taxa C lades CT Taxa Classically, taxonomists reach a consensus on ap­ CTG propriate ranks for taxa based on a broad knowledge CG GT of diversity, judgement of degrees of structural disparity, and respect for sensible traditions of what degree of disparity merits a particular rank. Though ranking has a subjective element, it is a very useful device for helping the human mind quickly ap­ G rades preciate the relative position of nested taxa within Fig. 3. The non-cquivalence of clados and taxa and the the taxonomic hierarchy and, roughly, the relative non-exclusiveness of cither with grades. Biological groups magnitude and significance of the differences be­ may be simply clades, grades or taxa, or any two or three tween them. It is a mistake to denigrate the process of these at the saíne time. Some well known groups, such of ranking because it cannot be objectively pro­ as and eukaryotes, arc simultaneously taxa, clades, and important grades of organization (region grammed into a computer. CTG). Some clearly definable clades should not be made Human judgement of significance and reasoned taxa, partly because they do not constitute clearly defined choice between alternatives are required throughout grades of organization (region C) (e.g. opisthokonta. i.e. science, as Singer (1931: 121-123) perceptively the cladc comprising Fungi, Animalia and Choanozoa emphasized; the Baconian error of thinking that one (Cavalier-Smith, 1987a); or ncokaryotes, the clade should be able to make important discoveries or consisting of all descendants of the most recent common sensible classifications merely by passing facts ancestor of Neozoa). Yet other clados are grades of 'through a sort of automatic logical mill’ (Singer, organization, but for other reasons arc not made taxa 1931: 121) is unfortunately widespread amongst (region CG; e.g. M ctakaryota, which was treated as a cladists. taxon in the eight kingdom system, but not in the six- Even if we were given the true phylogenetic kingdom system for reasons discussed in the text). relationships of all organisms, creating a balanced Paraphyletic taxa (regions T and GT) arc not clades; but they may be grades of organisation of greater (e.g. classification would require careful taxonomicjudge- Bacteria, Archezoa) or lesser (e.g. Reptilia, Amphibia) ment in delimiting and ranking taxa: classification importance (region GT) or they may not be grades of should not be based on a dogmatic approach that organization at all (region T) cither because they have prohibits paraphyletic taxa or a simple algorithm or secondarily become heterogeneous (e.g. the heterokont rule like that of Hennig (1966), which stated that the ‘algal’ phylum by multiple secondary losses categorical rank of a taxon should be determined by ofplaslids) or because they are too similar lo related taxa its biological age. The latter is practically impossible, to be called separate grades of organization (as in as we do not know', and may never know precisely, paraphyletic species ancestral to allopolyploids, e.g. where the root of the tree of life is located, and there Spartina anglica a n d S. maritima). Though many holo- are numerous places in the tree with so much phvlelic taxa are grades (region CTG , e.g. Tracheophyta) uncertainty about branching order that attempts to some arc not, because of secondary changes that have apply the rule would be arbitrary and less stable converted some of their members to a drastically different grade (region CT; e.g. Animalia, where the basic than the classical method of ranking by taxonomists. triploblaslic body plan has been totally lost by Myxozoa; If, for the sake of argum ent, ive knew that the root of Heterokonta where the Opalinata have lost the ciliary the tree of life was within the , which relroneme synapomorphy that makes other heterokonts is not unreasonable given the palaeontological an important grade of organization). Finally many evidence (Schopf, 1993), then It would be absurd to im portant grades are neither clados nor taxa (region G) ; rank certain cyanobacteria) subgroups more highly they may be paraphyletic groups (e.g. eubacteria, than the kingdom Animalia or the superkingdom , fish) or polvphylctic ones (e.g. protists, Eukaryota, merely because they diverged earlier. zooilagcllates, amoebae). All seven types of biological The morphological and chemical uniformity of the groups have valuable roles lo play in the description and cyanobacteria is so marked that they should be analysis of organismic diversity; it is a simplistic and ranked only as a phylum, and their subgroups lower harmful error to attempt to restrict the scientifically appropriate use of any of them by naive dichotomies into still irrespective of their probably immense an­ allowable and forbidden types of group, and still worse to tiquity; as in most other groups their molecular trees pretend that the three types that are not clades are not reveal so many rapid bush-like radiations (Turner, even groups al all. 1998) that their branching order would be an almost A revised six-kingdom system o f life 215

worthless guide to ranking, even were it philo­ (1) The concept of a bacterium (synonym: sophically desirable, which it is not. Significant prokaryote) evolutionary change is highly non-clock like : because of periods of stasis or minimal change despite branching, and an erratic distribution of major The concept of bacteria was considerably refined by innovations across the tree of life, there would be Picken (1960), Stanier (1961) and Stanier & Van no classificatory value in ranking taxa by their Niei (1962). These authors and Echlin & Morris presumed antiquity instead of by the classical (1965), Allsopp (1969} and Stanier (1970) con­ criterion of their structural disparity. trasted the structure of bacteria with that of eukaryotes in 12m ajor respects: (1) the fundamental differences in the structure and function of bacterial (4) The need to weigh and integrate flagella and eukaryotic cilia (a term which, for phylogenetic evidence from diverse sources clarity, embraces also eukaryotic ‘flagella’; Cava­ lier-Smith, 1986 A) ; (2) the location of respiratory Systematics is, par excellence, a synthetic science chains in the cytoplasmic (surface) membrane rather and needs both generalists and specialists. It is a than in mitochondria; (3) the location of photo­ commonplace that excessive reliance on a single line systems in either the cytoplasmic membrane or free of evidence is dangerous and that not all data are of thylakoids rather than chloroplasts; (4) the usual equal value; but judgement of their appropriate presence of a peptidoglycan cell wall; (5) the absence value is both difficult and controversial. Congruence of an internal cytoskeleton of actin microfilaments between different types of evidence is the most and tubulin-containing microtubules; (6) the ab­ im portant criterion for assigning weight to them. But sence of cytoplasmic motility mediated by ATPase there is no magic formula for doing so and the molecular motors such as myosin and dynein; (7) creation of a macrosystem of life is both an art and organisation of the chromosome as a nucleoid in the a science (Minelli, 1993} - not an art that passively cytoplasm, not as a nucleus; (8) absence of an reflects nature, but one which creates a simplified endomembrane system (endoplasmic reticulum, picture of its underlying pattern to further its Golgi complex and lysosomes) ; (9) absence of appreciation by humans. Like any science, macro- phagocytosis, endocytosis and exocytosis; (10) DNA systematics advances by trial and error: by intuition segregation by association with the cell surface, not and imagination as well as by careful observation, a mitotic spindle; (11) recombination by parasexual experiment, reasoning and critical discussion. I processes, not syngamy and meiosis; (12) inability to w'elcome criticisms of the present system, which harbour cellular . attempts to take some account of all types of Each of these twelve character differences between evidence, but deliberately gives them unequal bacteria and eukaryotes is of profound evolutionary weight. I have attempted to give reasons for major importance and significance. Collectively, they are decisions, but space does not permit this for every of such overwhelming significance compared with d e ta il. other character differences among organisms, that Stanier, Doudoroff & Adelberg (1963) were entirely correct in proclaiming the bacteria/eukaryote di­ chotomy as the most profound and significant in the III. IN DEFENCE OF BACTERIA: THE SOLE living world. Nothing that has been discovered since PRIMARY KINGDOM OF LIFE then has invalidated their thesis that it outweighs in importance the differences between the eukaryotic It is highly desirable that we keep the long kingdoms and equally strongly outweighs the dif­ established group ofBacteria as a major taxon in the ferences between the various groups of bacteria. classification of life. W hether Bacteria arc ranked as Since then, my own analyses (Cavalier-Smith, a kingdom or a superkingdom is less im portant than \981fe, 1987r, 1991o,i>) have added many more the retention of both the name and the concept of such differences of genetic and cellular organisation, bacteria in their classical meaning. However, I shall ranging from the presence of DNA gyrase in bacteria argue that the advantages of ranking Bacteria as a but not in eukaryotes to the absence of spliceosomes, kingdom rather than a superkingdom outweigh the peroxisomes and hydrogenosomes in bacteria. The disadvantages. Before explaining my present views, I total list now extends to about 30 major differences need briefly to review some im portant earlier ideas. (Cavalier-Smith, 1991a, b). 21 ö T. Cavalier-Smith

distinctiveness of Archaebacteria was, however, (2) Changing views on Archaebacteria historically useful in that it encouraged a horde of The characterization of Archaebacteria and of very competent bacteriologists and molecular biolo­ their substantial differences from Eubacteria (Woese gists to study them in the hope that they might & Fox, 1977a; Woese, 1987) has been of profound unearth some really radical differences in genetic importance for our understanding of bacterial organization from that of E . coli and eukaryotes. diversity. So also have been the methods of rRNA W hat this enormous enterprise has shown, how­ phylogenetics pioneered by Woese (1987; see also ever, is that Archaebacteria are not really very Olsen & Woese, 1993) for the progress of phylo­ different at all from E . coli in the organization of genetic knowledge of both bacteria and eukaryotes. their genes (e.g. in opérons) and , or their 1 do not wish to minimize the great importance of replication, transcription and machinery, these very positive contributions to science, but I as Keeling, Charlebois & Doolittle (1994) have must emphasize, as I have rather more mildly before clearly accepted. The first complete se­ (Cavalier-Smith, 1986c), that the systematic im­ quence of eubacteria and an archaebacterium give portance of the distinction between archaebacteria added force to the view that the genetic organization and eubacteria has been greatly exaggerated, a view of all bacteria is fundamentally the same, and strongly supported by M ayr (1990). Conversely, the radically different from that of eukaryotes (Edgell el very much greater importance of Stanier's (1961; al., 1996). Admittedly there are some differences, Stanier & van Niei, 1962) concept of bacteria for but these (except for some aspects of transcription) : the two words became synonyms with are ones of relatively small detail, mainly of interest the acceptance of Stanier’s term cyanobacteria foT to the specialist moleculai biologist rather than the what were previously known as blue-green algae), general bacteriologist, and these special properties of and of the Bacteria/eukaryote distinction, has been archaebacteria are mostly shared with eukaryotes, repeatedly attacked by rhetoric and assertion rather rather than indicative of a ‘third form oflife’. These than by careful argument. A long talk with George differences pale into insignificance in comparison Fox 10 years ago made it clear to me, however, that with the immensely more profound differences when he and Woese first started to criticize the between the genetic organization of eukaryotes and concept of a prokaryote (Woese & Fox, 19776) they bacteria, which Stanier (1961, 1970) was largely did not really have Stanier’s ideas (1961 ; Stanier & unaware of, but which I have discussed in detail Niei, 1962) in mind at all, but a naive view then elsewhere (Cavalier-Smith, 19816, 1987c, 1991a, 6, widespread among molecular that Escher­ 19936). This conclusion clearly refutes the idea that ichia coli was a typical prokaryote and that by Archaebacteria and Eubacteria arose independently studying its the whole of bacterial from a crude precellular progenote with poorly biology would be revealed. This naive view clearly developed genetic mechanisms (Woese & Fox, deserved to be criticized. W hat was most unfortunate 19776). T hat idea was based on the huge differences about the way that the criticism was made, though, between the two taxa revealed by 16S rRNA was that it was done as a blanket criticism of the oligonucleotide cataloguing. The way in which this perfectly sound basic idea of what constituted a technique severely exaggerates the quantitative prokaryote, rather than of the naive views of differences was trenchantly criticized by Hori, Itoh molecular biologists about the matter. Molecular & Osawa (1982), and the validity of their criticism biologists, being chemically rather than morpho­ has been confirmed fully by total rRNA sequences. logically oriented, never took much notice of However, even at the time it was proposed it was Stanier’s concepts (1961, 1970; Stanier & Niei, probably obvious to any cell biologist, but curiously 1962), which were mainly cell biological. Instead was not to most molecular biologists, that the idea they pictured a prokaryote in terms of such mol­ that eubacteria, archaebacteria and eukaryotes had ecular genetic features as operon structure and evolved independently could not possibly have been prom oter organization in which E . coli was known to correct. Long before then it was known that differ from eukaryotes. Woese and Fox ( 1977¿) were eukaryotes and eubacteria share so many features of correct to criticize the assumption that all bacteria cell organization and that their latest would necessarily have these features, but failed to common ancestor must have been a highly developed realize that such details played no role whatever in cell with several thousand genes, and highly de­ the basic organismic and cell-biological definition or veloped DNA replication machinery (Edgell & concept of bacteria. Exaggerating the biological Doolittle, 1997), and therefore could not have been A revised six-kingdom system o f life 217 a ‘ progenote ’ with crudely-developed replication, data, morphology, and non-sequence chemical data DNA repair, transcription and translation machin­ are complementary types of scientific evidence that ery (Woese & Fox, 1977¿). Belatedly, several authors have to be integrated in a balanced way for sound have become aware of this (e.g. Ouzonis & Kyrpides, systematic decisions. It is historically incorrect to 1996). imply that before rRNA sequencing there was no Woese's (1983) statement that it was impossible phylogenetically sound (Woese, for eukaryotes to have evolved from bacteria was a 1987; 1994): it was even less correct to imply that gross overexaggeration. I previously attempted to this was true of in general, since for explain in detail how a bacterium could have evolved eukaryotic microbes there are a wealth of good into the first eukaryote, and why Woese’s criticisms morphological characters and over a century of good (1983,1987 ; see also 1994) of the concept of bacteria phylogenetic studies and creation of durable higher are mistaken (e.g. Cavalier-Smith, 1987c, 1991 a, b). taxa. Even in bacteria it is a serious mistake to The evolution of rRNA is im portant, but one cannot dismiss morphology altogether (Woese, 1987) ; mere understand bacterial evolution by focusing mainly differences in cell shape are indeed often relatively on just one molecule (e.g. Woese, 1987). Bacteria are trivial, but ultrastructural morphology in bacteria is cells and organisms, and it is these biological entities exceedingly im portant and phylogenetically inform­ (not single molecules) that systematists must classify. ative. Long before any sequences became available Woese (e.g. 1987) has correctly criticized earlier there were several phylogenetically sound groupings bacterial systematics for its phenetic rather than of bacteria based on morphology, either alone or in phylogenetic approach, but this earlier approach combination with chemical characters. was as much through necessity as choice. Ribosomal The three most obvious such groupings are RNA has added a wealth of new characters without Cyanobacteria, and Gram-positive bac­ which a phylogenetic approach would have been teria, all established in the nineteenth century, and much more limited. But to understand cell evolution all supported much later first by electron microscopy we need to consider very much more than just and then eventually also by rRNA phylogeny (for a rR N A . summary of the bacterial classification proposed I have felt obliged to go into this history in some here see Table 2). The phylum Cyanobacteria is detail because the very early ideas of Woese & Fox exactly the same in definition and circumscription (1 9 7 7 a , b) about the independent and early ‘pri­ as the Myxophyta of Cohn (1875): rRNA only m ary’ divergence of archaebacteria, eubacteria and confirmed what we already knew. For leptospiras eukaryotes from a ‘progenote’, though not now and spirochaetes, Stackebrandt & Woese (1981) accepted even by the authors themselves, and the using trees based on rRNA oligonucleotide cata­ thoroughly mistaken criticism of the concept of loguing suggested that they were not related. I never bacteria and prokaryotes, have played a larger role accepted this, because the location, uniquely in in determining the present views of many biologists bacteria of the flagella in the periplasmic space on the status of archaebacteria, than has a critical seemed such a complex and unusual morphological assessment of the overall evidence now available to character that it was highly unlikely to have evolved us. I wish, however, to emphasize that my criticisms twice. Therefore, my first bacterial classification of these particular early views in no way detracts (Cavalier-Smith, 1987a) treated them as a single from the importance of the concept of archaebacteria phylum. Full rRNA sequences agree with the for understanding cell evolution and biological morphological conclusion that this phylum (Spiro- d iv ersity . chaetae) is monophyletic (Olsen, Woese & Over- beek, 1994; Embley, H irt & Williams, 1995). This highlights the point often made by systema­ tists, but still not fully appreciated by some molecular (3) The importance of cell structure in biologists, that molecular characters are not in­ bacterial classification herently superior to morphological ones. There are Though rRNA sequencing has quite clearly caused a strong molecular characters (e.g. full 16s rRNA revolution, largely most beneficial, in bacterial sequences) and weak ones, just as there are strong systematics, it is a great pity that it has sometimes morphological characters (e.g. the periplasmic lo­ been associated with a neglect (and still worse a cation of flagella and the structure of positive disparagement) of morphology, which re­ cyanobacteria! thylakoids) and weaker ones (e.g. mains crucial for bacterial systematics. Sequence ro d s versus cocci). What is still sorely needed in 218 T. Cavalier-Smith

T a b le 2. ('.lanification o f the kingdom Bacteria and its IO phyla

Subkingdom1. Negibacteria* Cavalier-Smith 1987 [outer membrane present; acyl ester lipids; murein wall between the two membranes usually present; small recognition particle RNA (based on only at present) ; in content nearly the same as the former phylum ]. Infrakingdom 1. Lipobacteria* infrak. nov. (murein cell wall ; outer membrane of phospholipids; lipopolysaccharide absent; no flagellar shaft outside outer membrane). SuperphylumJ 1. * Cavalier-Smith 1982 (as an infrakingdom) slat. nov. (murein walls with ornithine; diaminopimelic acid, cytochrome ao3, and RuBisCo absent; flagella absent; small citrate synthetase). or PhylumJ 1. Heliobacteria Cavalier-Smith 1987 (anaerobic photosynlhcsisers unable to fix C 0 2, bacterioehlorophvll g a n d g ; no chlorosomes or h cytochromes). Phylum 2. Hadobacteria Cavalier-Smith 1992 cm. (thermophiles or radiation resistant; if photosynthetic fix cos). Subdivision or subphylum]: 1. Chlorobacteria§ Cavalier-Smith 1992 (aerobic photosynthctic thermophilic gliders with bactcriochlorophyll c a n d b cytochromes; e.g. Heliolhrix, Chloroflexus). Subphylum 2. Deinobactcriaf Cavalier-Smith 1986 (non-photosynthetic; thermophilos or radiation resistant with thick walls, e.g. , Thermus ). Supcrphylum 2. Endoflagcllata Cavalier-Smith 1992 (flagella in periplasmic space). Phylum 1. Spirochactac Ehrenberg 1855 stal. nov. Subphylum 1. Euspirochactae subphyl. nov. (murein walls with ornithine, e.g. ). Subphylum 2. Lcplospirae subphyl. nov. (murcin walls with diaminopimelic acid, e.g. Leptospira) Infiakiugdom 2 . Glycobacteria infrak. nov. (outer membrane with phospholipid in the inner leaflet and (¡polysaccharide in the outer leaflet of the bilayer; RuBisCo present; flagella pass through outer membrane). Superphylum 1. Pimelobactcria* Cavalier-Smith 1992 (as infrakingdom) stal. nov. (murein wall with diaminopimelic acid). Phylum 1. Cavalier-Smith 1987 (sphingolipids; flagella absent, usually glide) Subphylum 1. Chlorobibacteria subphyl. nov. (photosynthctic anaerobes with chlorosomes: Chlorobiaceae, e g. Chlorobium). Subphyltim 2. Flavobacteria (non-photosynthetic, e g. Flavobacterium, Cytophaga). Phylum 2. Eurybacteria* phyl. nov. (without sphingolipids or photosynthesis; phylogenetically distinct from proteobacteria; monophyly of this phylum is more uncertain than for any other bacterial phyla, as clear synapomorphies not identified). Subphvlum 1. Sclenobactcria* Cavalier-Smith 1992 (as phylum) stat. nov. (non-fusiform ; often , e.g . Selenomonas, Sporomusa)■ Subphylum 2. subphyl. nov. (fusiform, non-flageltale; Fusobacieiium. Leptotrichia). Subphylum 3. Fibrobactcria subphyl. nov. ( Fibrobacter ). Phylum 3. Cyanobacteria Stanier 1973 (oxygenic photosvnthcsizcrs with chlorophylla). Subphylum 1. Gloeobacteria subphyl. nov. (no thylakoids; phycobilisomes on cytoplasmic mem brane; Gloeobacter). Subphvlum 2. Phycobactcria subphyl. nov. (with thylakoids, e.g. Anabaena. Jsostoc, Prochloron). Phylum 4. Proteobacteria Stackebrandl et al. 1986 (without sphingolipids; phylogenetically distinct from Eurybacteria). Subphylum 1. Rhodobacteria Cavalier-Smith 1987 (as phylum) stat. nov. (pholosynthetic purple bacteria and their colourless derivatives; w ithout chlorosomes, phycobilisomes or thylakoids; photosynthetic machinery in cytoplasmic membrane invaginations; bactcriochlorophyll r andd and purple carotenoid in photosynthetic species; this subphylum and its three major subgroups arc delimited primarily by ribosomal RNA similarity as clear synapomorphies have not been found). Infradivision or infraphylum]: 1. Alphabacteria Cavalier-Smith 1992 ias class) slat. nov. (with or w ithout photosynthesis, e-g. Rhodospirillum, Rickettsia, Agrobacterium). Infraphylum 2. Chrom alibacteria infraphyl. nov. (with or without photosynthesis; /?- and y- proieobarlcria, e.g. Escherichia, Haemophilus, Spirillum, Chromatium). Subphylum 2. Thiobacteria subphyl. nov. fS - and e-protcobaclcria; non-photosynthetic, often sulphur- dependent, e.g. Desulfovibrio, Thiovulum, Bdellovibrio, M yxobacteria, e.g. M yxococcus]. Supcrphylum 2. Planclobactcria supcrphyl. nov. (wall of protein, not murein peptidoglycan). Phylum Planclobactcria Cavalicr-Smilh 1987 (Planctomycetales and ). Subkingdom 2. Unibacteria new subkingdom (diagnosis: bacteria with single cytoplasmic membrane only, but no outer m em brane; large signal recognition particle RNA). A revised six-kingdom system o f life 219

T a b le 2. [corii.)

Infrakingdom 1. Posibacteria* Cavalier-Smith 198? stat. nov. (acyl ester lipids; murein widespread). Phylum Posibacteria* Cavalier-Smith 1987. Subphvlum 1. Teichobactcria subphyl. nov. (cell walls, if present, thick and with teichoic acids; includes and M ollicutes; paraphyletic Firmicutes here abandoned as a formal taxon and subdivided between Endobacteria and ). Infraphylum Endobacteria infraphyl. nov. (spores endosporcs; DNA low in guanine and cytosine; classes Clostridea Cavalier-Sm ith 1982, e.g. Bacillus, and , e.g. ). Infraphylum 2. Actinobacteria Margulis 1974 (as phylum) slat. nov. (spores cxospores; DNA high in guanine and cytosine; e.g. , Streptomyces). Subphylum 2. Togobactcria Cavalier-Smith 1992 (as phylum) stat. nov. (teichoic acids absent; murein wall very thin; external non-lipid toga; obligatcly anaerobic thcrmophiles-. Thermotogales andA quifex',. Infrakingdom 2. Archaebacteria Woese & Fox 1977 stat. nov. (isoprenoid ether lipids; murcin absent). Phylum Mendosicutes Gibbons & M urray 1978. Subphylum 1. Euryarcheota Woese, Kandier & Whcclis 1990 stat. nov. (energy metabolism various; not dependent on elemental sulphur). Infraphylum 1. Halomebacleria Cavalier-Smith 1986 (mclhanogens or extreme halophilcs, e.g. Heliobacterium, Methanospirillum). Infraphylum 2. Eurylhcrm a infraphy. nov. (non-mcthanogenic, non-halophilic thermophilos; Thermoplasma, Thermococcales, Archaeoglobales). Subphylum 2. Sulfobaclcria Cavaiier-Smilh 1986 stat. nov. (syn. Crcnarchcota Woese, Kandier & Wheelis 1990 : energy metabolism depends on elemental sulphur; and ; Jim Black Pool thermophilos: e.g. Sulfolobus, Pyrobaculum).

This classification is modified from that in Cavalier-Smith (1992a), where its cladistic basis is explained in more detail. To save space diagnoses in Tables 2-7 are given only for new taxa. * Probably paraphyletic. I Almost certainly paraphyletic. J All the phyla here should be treated as divisions for formal bacterial nomenclature, and the superphyla as superdivisions, subphvla as subdivisions, and infraphyla as infradivisions. For uniformity with the other kingdoms of life, I have used phylum rather than division in the present paper. The use of division rather than phylum is an his­ torical relic of the origin of the Bacterial Code of Nomenclature from the Botanical Code rather than from the Zoological Code, which impedes a unified approach to biological nomenclature and systematics. As the Botanical Code now allows phyla and states that phylum and division are of equivalent rank, the Bacterial Code ought to be similarly revised. W hen the Unified Code of Biological Nomenclature, now being planned for the next century (Hawksworth, 1995), is introduced, I hope that it will adopt phylum, not division, for all organisms and that it explicitly recognizes subkingdoms, infrakingdoms, superphyla, subphyla and infraphyla so as to fix their relative rank, and also either superkmgdoms or empires or both. By using phylum not division, one is free to use ordinary English terms like subdivision in a general sense with no connotation of rank or risk of their being confused with the taxonomic category subphvlum. Haeckel's (1866) name M onera is an unsuitable name for bacteria as it was invented for hypothetical non-nucleate amoebae, which do not exist, and is much less widely known than Bacteria. § Originally treated as separate phyla (Cavalier-Smith, 1992û); for molecular evidence for their monophylv see Olsen, Woese & Overbeck (1994).

bacterial systematics, and has not been provided by microscopy showed that the vast majority of Gram- the sequence-oriented school, is the development of a positives had only a single bounding membrane, stronger tradition, like that established over two whereas the vast majority of Gram-negatives were centuries in eukaryotic systematics, of the critical bounded by two separate membranes. I have long weighing of all available characters, whether mol­ argued that this morphological difference is of very ecular, chemical or sequence. Over-emphasis on one profound cell biological and evolutionary import­ molecule and neglect of important morphological ance, and therefore coined the word Posibacteria for evidence is poor systematic practice. all eubacteria with a single membrane, because their The Gram-positive group, as defined by light most numerous members are the true Gram-positive microscopy, was slightly heterogeneous. Electron bacteria (Cavalier-Smith, 1987r). The taxon Posi- 220 T. Cavalicr-Smith bacteria was originally both a phylum and a valid name Mendosicutes (Gibbons & Murray, subkingdom but is now ranked as both a phylum 1978). and an infrakingdom. Posibacteria is not a synonym for true Oram-positive bacteria (formally Firmicutes (4) Characters important in the high-level or Firmibacteria), as is sometimes incorrectly classification of Bacteria thought, because it also includes two other taxa: these are (Mollicutes : another m orph­ Archaebacteria, like Posibacteria, have only a single ologically based group confirmed by rRNA se­ bounding membrane. I shall therefore henceforward quences - both sequences and membrane chemistry refer to both groups collectively as Unibacteria. showed that the archaebacterium Thermoplasma Because of the differences in protein and lipid acidophilum was originally wrongly included, so targeting that it implies, I think that the difference morphology had to be supplemented by this extra in membrane number between Negibacteria and evidence to refine the group) and Thermotogales. Unibacteria is the most important morphological Sequencing showed that mycoplasmas evolved from difference of all within the Bacteria (Cavalier-Smith, the branch of Firmicutes with DNA of low guanine 1987c, c; 1991 a, b, 1992c). By contrast, the differ­ and cytosine content by the loss of the murein ences in morphology between Archaebacteria and peptidoglycan wall (Woese, Stackebrandt & Eubacteria are trivial or non-existent. The cells of Ludwig, 1985). This molecular sequence demon­ Archaebacteria and Posibacteria are organized in stration that Firmicutes and Mollicutes together are fundamentally the same way; there are no known monophyletic is congruent with the morphological morphological differences between them by which fact that both groups have a single membrane, not all archaebacteria can be distinguished. They have two as in all other eubacteria. Thus the singleness of morphologically indistinguishable flagella and gas the membrane and the acyl ester character of its vacuoles. Archaebacteria are undoubtedly bacteria lipids are more important than the presence or by all 12 of Stanier’s (1961, 1970) criteria listed at absence of the cell wall that was emphasized in the the beginning of subsection (1). Their genetic earlier classification (Gibbons & M urray, 1978). systems are very similar to those of Posibacteria To contrast all the other eubacteria that are (Doolittle & Brown, 1994; Keeling, Charlebois & bounded by two distinct lipid membranes with Doolittle, 1994), and their cell-division apparatus is Posibacteria, I named them Negibacteria (Cavalier- fundamentally similar (Wang & Lutkenhaus, 1996). Smith, 1987 r). Negibacteria is not an exact synonym The complete sequence of the first archaebacterial for the vernacular term Gram-negative bacteria, g e n o m e {Methanococcus jannaschii: B ult et a i , 1996) in because it includes a few bacteria such as the genus comparison with those of the eubacteria Haemophilus Deinococcus that are Gram-positive because they have influenzae (a negibacterium : Fleischmann et al., 1995) a thick wall between the two membranes, rather a n d Mycoplasma capncolum (a posibacterium Fraser et than outside the single membrane as in the majority al., 1995) confirms that both eubacteria and archae­ of Gram-positives. However, it has the same cir­ bacteria are fundamentally similar in genome or­ cumscription as the division Gracilicutes (Gibbons ganization and content, and share nearly 1000 & Murray, 1978), but the higher rank of sub­ different genes. M any, if not most, of the genes that kingdom. Therefore it could be argued that it was, cannot at present be homoiogized across the eubac- strictly speaking, an unnecessary new word, like the terial/archaebacterial divide may simply be rather term Cyanobacteria, a fifth synonym for Myxophy­ rapidly evolving ones, rather than genuinely unique ceae, Cyanophyceae, M yxophyta and Cyanophyta. to either group. As Edgell et al. (1996) rightly stress, However, as Gracilicutes was of much lower rank, the most striking thing about the eubacterial and not in wide currency, and there was no informal archaebacterial genomes is how similar they are; as vernacular equivalent, I thought it useful to create a these authors explain, the apparently high fraction subkingdom name that would be euphonious in both of archaebacterial genes not previously present in the vernacular (as negibacteria) and formally (as databases is almost certainly an artefact of the Negibacteria), and which would contrast the group present low representation of archaebacterial sequen­ directly with Posibacteria by using the prefix Negi- ces compared with those of eukaryotes and eubac­ and emphasize their bacterial character by using the teria, rather than a sign of the uniqueness of suffix -bacteria. The importance of euphony is archaebacteria as Bult et al. ^ 1996) somewhat illustrated by the widespread use of the name misleadingly suggested. As more sequence becomes Archaebacteria in preference to the first formally available for the crenarcheote archaebacterium A revised six-kingdom .system o f life 221

Sulfolobus solfataricus (S ensen et al., 1996) this artefact bacteriologists had no tradition of using categories of will be reduced. high rank such as phyla; virtually all classifications If one wished to divide Bacteria into two kingdoms used oniy classes, orders and families. There were no based on morphology then I would argue that phyla, subphyla, or subkingdoms at all: nothing Negibacteria and Unibacteria would be a better between classes (and precious few of them) and the choice than Eubacteria and Archaebacteria. kingdom Prokaryota. After the recognition of In what non-morphological ways do archae­ Archaebacteria four phyla were introduced : Gracili­ bacteria and eubacteria differ that might be suf­ cutes, Firmicutes, Mollicutes (formerly a class), and ficient to justify their separation into separate Mtndosicutes (Archaebacteria) by Gibbons & kingdoms? They differ in wall chemistry. The M urray (1978). Then suddenly, on the basis just of absence of murein in archaebacteria has often been rRNA sequence divergence, a dozen or more stressed but it is also absent in the posibacterial separate bacterial kingdoms were suggested seriously mycoplasmas and in the negibacterial Plancto- (Woese, 1987), and still seem to be favoured by bacteria. Probably, as I have previously argued Olsen, Woese & Overbeek (1994), but have for­ (Cavalier-Smith, 1987a, c), this reflects three in­ tunately not been formally adopted by bacterial dependent losses of murein : but this negative taxonomists. From a system where major groups character clearly would not even justify their such as spirochaetes and Cyanobacteria were given separation as a subkingdom, still less a kingdom. much too low a rank in comparison with eukaryotic More important is the presence in archaebacteria, taxa, there was a dramatic jum p to excessively high but apparently not in eubacteria, of N-linked ranking of the major bacterial taxa. glycoproteins, a character that is shared with The overall system of life must be well balanced if eukaryotes and which I used to argue for a sister it is to serve the needs of the general biologist as well group relationship between archaebacteria and as those of the specialist in some particular group. It eubacteria (Cavalier-Smith, 1987c). Given that O- is im portant for systematics and biology as a whole linked glycoproteins are present in a few eubacteria, that ranking within bacteria does not get drastically as well as archaebacteria and eukaryotes, the out of step with that of eukaryotes. magnitude of the innovation is not immense. There The widespread ranking of Eubacteria and is also a most interesting difference in the chemistry Archaebacteria as separate kingdoms has never been of the shaft of the in archaebacteria really critically discussed. The main justification (F e d o ro v et a i , 1994). But none of these differences usually assumed for such a high rank is the large is so profound individually or collectively to justify a difference in 16S rRNA sequence between these two separate kingdom. The well known difference in groups. This originally so impressed Woese that he membrane lipid chemistry (isoprenoid ethers in thought that archaebacteria must be a radically new archaebacteria and acyl esters in eubacteria) is the form oflife (Woese & Fox, 1977a). But what does the only other im portant qualitative chemical character large difference in this one molecule really amount that clearly differentiates archaebacteria from cubac- to biologically? The molecule is over 50% identical teria. I have long considered that this difference is between the two taxa and functions in much the sufficiently important, coupled with the internal same way in both groups. There is no reason to think metabolic diversity of archaebacteria, to give the that most of the differences are of substantial taxon some sort of supraphyletic rank. But why not functional significance: they probably are mainly a subkingdom, infrakingdom or superphylum rather biologically meaningless chemical noise that has than a separate kingdom? The resemblances be­ accumulated by random mutation and genetic drift tween archaebacteria and eukaryotes in transcrip­ of neutral or quasi neutral sequences (Kimura, tion factors and some details of rRNA processing, 1983). As such, they are very important as phylo­ though supportive of a sister relationship between genetic markers, but do not constitute the sort of the two groups (i.e. of the clade : Cavalier- organismal characters properly used in the classi­ Smith, 1987c), involve insufficient differences from fication and ranking of higher taxa. Quantitative the eubacterial pattern to justify any higher rank differences in rRNA would be a very arbitrary basis than infrakingdom or superphylum. for deciding rank, because rates of rRNA evolution Bacteriologists and molecular biologists seem to vary considerably, sometimes drastically (up to have given rather little thought to the possibility of about 50-fold), in different lineages and do not using such taxonomic categories of intermediate correlate well with the occurrence of substantial rank. Indeed, before the advent of rRNA sequencing differences in body plan, w'hich have always been 222 T. Cavalier-Smith

(and ought to continue as) the primary basis for ating characters are ancestral and which are derived ; delimiting higher taxa. The differences between for greater weight in grouping is accorded to shared archaebacterial and eubacterial rRNA are com­ derived characters (synapomorphies) than to shared parable in magnitude (though actually slightly less ancestral ones. This brings me to the vexed question than) those that separate the two eukaryote phyla of rooting the tree of life, which is far more difficult Microsporidia and , which differ far than is usually thought and has not yet been more from each other morphologically, and in way unambiguously achieved. of life, than do the two bacterial groups, yet are now For over a century, most biologists have assumed both included in the kingdom Fungi (see that eukaryotes evolved from bacteria, not vice VII). versa, both because it was much more reasonable to Ribosomal RNA on its own provides no sound suppose that the structurally simpler bacterial cell reason to rank Archaebacteria any higher than a preceded the much more complex eukaryotic one, phylum. In fact, I consider that the rRNA differ­ and because, since the 1950s, that view has been the ences deserve less weight than do the membrane best interpretation of the now very extensive mi­ chemistry, the N-linked glycoproteins, and the crobial fossii record (Schopf & Klein, 1992). Woese presence of a single bounding membrane in deter­ & Fox (1977a), however, relying solely on the rRNA mining the rank of Archaebacteria. It is entirely cataloguing evidence and the simple, but invalid unwarranted to treat the two archaebacterial sub­ assumption of a molecular clock, proposed instead phyla as separate kingdoms; erecting a third that eukaryotes, archaebacteria and eubacteria were archaebacterial 'kingdom ’, , just on equally old and had evolved independently from the the basis of rRNA divergence, was most unwise, hypothetical progenote. For these three taxa, they especially as we know next to nothing about the coined the novel term Urkingdom to emphasize their phenotypes of the organisms in question (Barnset a l., hypothesis that the three groups were derived 1996). The discovery of these previously unknown independently as separate ‘prim ary’ kingdoms. For lineages, coupled with the uncertainty of the branch­ the next decade many molecular archaebacteriolo- ing order at the base of the archaebacterial rRNA gists accepted this idea overdogmatically, even tree and the known propensity of rRNA for highly though cell biological and palaeontological argu­ misleading non-clock like evolution (see discussion ments were always opposed to it. below on microsporidia) should discourage the use of The situation was radically changed not so much rRNA branching depth alone as a basis for sub­ by my detailed arguments about cell evolution dividing bacteria or any other organisms. W hen we (Cavalier-Smith, 1981a, b, 1987a, c), and the cla­ know more about the biology of the novel uncultured distic considerations that favoured a sister relation archaebacteria, it may be appropropriate to sub­ between archaebacteria and eukaryotes (Cavalier- divide archaebacteria into more than one phylum. Smith, 1987 c), but by molecular trees for duplicated But at present I favour a single phylum for all proteins (Iwabe et a l., 1989; Gogarten et al., 1989). archaebacteria. These clearly supported a sister relationship between The above comments, from a predominantly archaebacteria and eukaryotes and suggested that eukaryotic systematist with extensive experience in they had separated from each other a substantial ranking higher level microbial taxa, now entering time after the origin of life. If the topology of these the bacterial field more seriously than before, are trees if correct, they provide cladistic evidence that intended to be constructive, and should not be the eubacterial/eukaryote acyl ester lipids are the construed as personal criticisms of other biologists, ancestral type and the archaebacterial isoprenoid despite their critical character in places. ones are derived, as wras argued earlier (Cavalier- Smith, 1987a, c). If the eubacterial lipids are indeed the ancestral type, then their common possession by (5) Bacterial subkingdoms and Negibacteria and Posibacteria is not a very strong infrakingdoms argument for linking them together in the taxon Taking both morphology and chemistry into ac­ Eubacteria. Eubacteria are a paraphyletic grade, count, we have, in fact, not two but three major not a holophyletic clade as archaebacteria probably groups of bacteria that have to be grouped and are. ranked: Archaebacteria, Posibacteria and Negi­ The likelihood that eubacteria are paraphyletic is bacteria. In deciding groupings and ranks systema- not in itself a reason for rejecting the taxon; 1 have tists traditionally try to determine which difi'erenti- treated them as a subkingdom or kingdom for many A revised six-kingdom system o f life 223

years despite believing them to be paraphyletic. But first negibacterium. Infrakingdom I.ipobacteria con­ it should cause us to ask seriously whether the best tains the three phyla which lack lipopolysaccharide place for the primary subdivision of bacteria is that in their outer membrane; two of these lack flagella between archaebacteria and eubacteria or whether and have ornithine rather than diaminopimelic acid it might not be better to place it within the (DAPA) in the cross-linking peptide of their murein eubacteria between the Negibacteria and Posi­ walls (the photosynthetic Heliobacteria and the bacteria. It is really a question of whether one partly photosynthetic and partly heterotrophic weights more strongly the difference in number of Hadobacteria ), while the Spirochaetae (typical membranes or the differences in membrane chem­ spirochaetes, with ornithine-containing muramo- istry and N-linked glycoproteins. I consider the peptides and leptospiras with DAPA) have flagella difference in membrane number to be of more within the periplasmic space. Glycobacteria are profound evolutionary significance, because there subdivided into two superphyla according to have been a number of changes in membrane whether they have murein walls or not. The chemistry, such as the invention of sterols, but glycobacterial superphylum Pimelobacteria, un­ apparently only one changeover between two and changed in circumscription but slightly reduced in one bounding membrane (Cavalier-Smith, 1980, rank compared with my previous classification 1987a, c\ 1991a, b ). I therefore here treat Negi­ (Cavalier-Smith, 1992a), have murein walls con­ bacteria and Unibacteria as bacterial taxa of equal taining diaminopimelic acid and include the tw'o rank, and both Posibacteria and Archaebacteria of best-known negibacterial phyla, Cyanobacteria and lower rank but equal to each ocher. Proteobacteria, as well as the lesser known Sphingo­ W hat rank should be assigned to these three taxa? bacteria and Eurybacteria. Superphylum Plancto- As Negibacteria are sufficiently diverse in cell bacteria includes only the Planctomycetales and structure and/or fundamental biochemistry to merit chlamydias (in a single phylum, the Plancto- subdivision into several phyla (eight being adopted bacteria), which lack murein. I have suggested here), all three taxa must be ranked at a level above previously that the absence of murein in Plancto- that of phylum. While a reasonable case could be bacteria is derived; and the presence of murein in made for treating Negibacteria and Unibacteria as the common ancestor of Lipobacteria and Pimelo­ separate kingdoms, I prefer to treat them as bacteria was the ancestral condition (Cavalier- subkingdoms of a single kingdom Bacteria, for Smith, 1987a). If this and my hypothesis that the basically the same reasons given below for the lipobacterial lack of lipopolysaccharide is ancestral kingdom Protozoa. It provides a simpler and more are both correct, then to understand the early balanced system for general reference purposes, and evolution of negibacteria we need to focus attention maintains a good correspondence between the on the little-studied Lipobacteria. If my view that kingdom name and the vernacular name ‘bacteria' the root of the tree of life should be placed within the that has been in perpetual use for over 150 years and negibacteria is also correct, this means that to is well known to the layman. If Unibacteria is a understand the earliest divergences in the history of subkingdom, it is most appropriate to make Archae­ life the study of Lipobacteria will be much more bacteria and Posibacteria mfrakingdoms. The pro­ important than that of Archaebacteria. It is, of posed broad classification of bacteria and its place course, possible that the absence of lipopoly­ within the six-kingdom system are summarized in saccharide, flagella, DAPA and RuBisCo in eobac- Table 1 ; the major taxa comprising each infra- teria is derived (or derived in some but not other kingdom are indicated in Table 2. taxa), but until there is definite evidence for this we The subkingdom Negibacteria is here divided into should take seriously the possibility that this was the two infrakingdoms based on the chemistry of the ancestral state for all bacteria and regard the cell’s outer membrane. One of these I call Glyco- eobacteria as very good candidates for the earliest bacteria, as it has lipopolysaccharide in the outer diverging cells and thus of great potential evol­ leaflet of the bilayer of the outer membrane; the utionary significance. other, which lacks lipopolysaccharide and has The present classification of Bacteria into two phospholipid in both leaflets of the outer membrane, subkingdoms, each further subdivided into two I name Lipobacteria, and suggest this is the ancestral infrakingdoms, has a pleasing symmetry and bal­ condition since lipopolysaccharide synthesis is im­ ance. The primary subdivision is according to the mensely more complex than phospholipid synthesis, number of envelope membranes, and the secondary and is therefore unlikely to have been present in the subdivision of each relies on membrane chemistry. 224 T. Cavalier-Smith

Each of the four infrakingdoms differs not only in photosynthetic (Woese, 1979; Cavalier-Smith, mem brane num ber and or chemistry, but also in cell 1987a). The presence of chlorosomes in C.hlorofiexus wall chemistry or structure. Archaebacteria alone (a lipobacterium) and Chlorobiaceae (Glycobac­ have N-linked glycoproteins and sometimes pseudo- teria) suggests that the transition between the two murein; Posibacteria alone have teichoic acids, and infrakingdoms involved a green photosynthetic bac­ typically very thick m urein walls w ith peptide linkers terium with chlorosomes, thus supporting the idea of very variable chemistry; Glycobacteria have very that these neglected bacteria are of pivotal evolu­ thin murein layers always with DAPA in the tionary significance (Cavalier-Smith, 1987a). crosslinking peptide; Lipobacteria have usually thin, Whether the ranking of Archaebacteria, Posi­ but sometimes thick, murein walls typically with bacteria, Eobacteria and Lipobacteria as infra- ornithine in the linker. Thus, the special cytoplasmic kingdoms and the abandonment of the taxon membrane chemistry of archaebacteria is given the Eubacteria will be accepted by others, only time will same taxonomic weight as the special outer mem­ tell. However, I hope that the reader will agree that brane chemistry of the glycobacteria. Isoprenoid it is based on a more careful consideration of what ethers and lipopolysaccharides are the synapo- weight should be given to the major differences morphics that can be used to define these two within Bacteria, than the suggestion that archaebac­ infrakingdoms most clearly. The four-infrakingdom teria and eubacteria be ranked equally as domains system groups the 10 bacterial phyla recognized here (Woese, Kandier & Wheelis, 1990). These are (a slightly lower num ber than that more tentatively merely unnecessary new names for the earlier suggested by Woese, 1987) into organismaliy more suggested Urkingdoms or ‘primary kingdoms’ of homogeneous higher taxa than does the archae- Woese & Fox (1977a) that were not well based. As bacterial/eubacterial dichotomy : the probably para­ I stressed previously (Cavalier-Smith, 1986c) there phyletic Eubacteria are so much more heterogeneous are no good reasons to rank the three taxa organismaliy than archaebacteria, that their split­ eukaryotes, archaebacteria and eubacteria equally ting into three taxa each of the same rank seems to as kingdoms; merely renaming them domains me a marked improvement and a more balanced (Woese, Kandier & Wheelis, 1990) does not lessen classification. the criticisms of equal ranking at all (Cavalier- The major phylogenetic assumptions behind the Smith, 19926); the new category is also six-kingdom system are explicitly laid out in Fig. 1. unnecessary given that we already have super­ Unlike traditional bacterial classifications, the pres­ kingdom and empire above the level of kingdom. ent system does not undervalue the diversity of Equal ranking of these taxa produces a grotesquely bacterial body plans in comparison with that of unbalanced system for living organisms as a whole. higher kingdoms; as Table 1 makes clear, the I he name eubacteria will, however, continue to be kingdom Bacteria has four major subdivisions of very useful to denote a grade, as are ‘’ rank infrakingdom or above, which is the same as in and ‘fish’ in or ‘alga’ in botany. The the kingdom Plantae and more than the kingdoms suggestion that eubacteria be renamed bacteria Fungi and Chromista, which have only two or three (W oese e! a t., 1990), was exceedingly confusing and respectively. Even Protozoa and Animalia, the most should not be followed (Cavalier-Smith, 19926). megadiverse kingdoms have, respectively, only six Some may object to my placing the Thermotogales and 11. In this system the taxa Archaebacteria and a n d A q u ifex in the Posibacteria on the grounds that Archezoa are each treated as one of the 30 major the rRNA tree does not provide evidence for such a forms of life; this sufficiently emphasizes their grouping. However, the rRNA tree neither provides distinctiveness - previously Woese and I were prob­ convincing evidence against such placing nor con­ ably both a little overenthusiastic in the use of the vincing evidence for any supraphyletic groupings unnecessarily high rank of kingdom. At the phylum within eubacteria, as has been made clear by level, Bacteria are the third-most diverse of the six Embley, Hirt & Williams (1995). On rRNA trees kingdoms. The fact that five out of the eight even all Teichobacteria often do not group together; negibacterial phyla (including more than one in for their two major subgroups (Endobacteria and each supcrphylum) are entirely or partly photo- Actinobacteria) often, but not always, separate from synthetic shows that the diversification of the basic each other (Olsen, Woese & Overbeek, 1994; photosynthetic machinery (Pierson & Olson, 1989) Embley, Hirt & Williams, 1995). It appears that the played a key role in early cell evolution and is eubacterial phyla (and sometimes also subphyla) consistent with the view that the first cells were underwent an almost simultaneous diversification so A revised six-kingdom system o f life 225 that molecular trees cannot correctly resolve their excessively long in comparison with the ideal branching order, or even, in the case of the situation for accurate tree-construction. The possi­ Teichobacteria, confirm or refute their monophyly. bility of such an artefact, in which the root is placed In such a circumstance, it is perfectly proper to base in the mid position rather than in the correct place, phylogenetic reasoning and classification solely on makes the interpretation of all gene duplication trees other evidence, which in some cases may be much studied up to now rather uncertain. stronger than the sequence evidence. I have argued It is important to stress that the uncertainty that this major eubacterial radiation most likely was regarding the position of the root of the tree of life the consequence of the origin of the first photo­ does not afTect the validity of the bacterial sub­ synthetic eubacterial cell and agree w'ith VVoese’s kingdoms and infrakingdoms proposed here. The (1979) hypothesis that the first bacterial cell was uncertainty simply means that we do not know photosynthetic. which of them are holophyletic and which are For reasons discussed in detail earlier (Cavalier- paraphyletic. If I am correct, then Archaebacteria Smith, 1980, 1987a, 1992a), I have argued that the are holophyletic and Posibacteria, Unibacteria, first cell was a negibacterium with two membranes, Negibacteria, Glycobacteria and Lipobacteria are and that Unibacteria are secondarily derived by the all paraphyletic; if, however, the tree of Iwabeet al. loss of the outer membrane. If this is true, Uni­ (1989) is correctly rooted, then Negibacteria would bacteria and Eukaryota together form a clade that become holophyletic, as would one of Glycobacteria may be referred to by the informal term uni- and Lipobacteria (depending on their branching membrana. The distinction between unimembrana order) ; the same taxa also become holophyletic if and negibacteria is profoundly important for cell Lake (1988, 1989) is correct and the root lies evolution and origins. The first cell must either have between the two archaebacterial subphyla; whereas been unimembranous, as is classically assumed (e.g. Archaebacteria become paraphyletic. Neither Goldacre, 1958; Hargraeves & Deamer, 1978), or grouping would make them polyphyletic. However negibacterial, as Blobel (1980) and Cavalier-Smith certain conceivable branching orders of the eubac­ (1987a) have suggested. Until we can establish terial phyla would make one or more of my taxa firmly which idea is correct, we shall continue to polyphyletic and thus necessitate a revision. Ac­ have a very shaky basis for understanding early cell curately determining their branching order will be evolution. It might be thought that a combination of the best test of monophyly of the major taxa in the th e Iw a b e el al. (1989) trees and the rRNA trees present classification. Both this and the fixing of the refute the idea that negibacteria were ancestral since position of the root may only be possible when a they appear to place the root between Archae­ complete genome sequence becomes available for at bacteria and Posibacteria, i.e. within Unimem brana. least one member of each of the nine eubacterial But I remain unconvinced that we really know phyla. As four are already published - Proteo­ where the root is yet, as do others (e.g. Forterre et al., b a c te ria [Haemophilus influenzae (Fleischmann et al., 1993; Doolittle & Brown, 1994). While the basic 1995); also E . coli a n d Helicobacter pylori\ C y a n o ­ logic behind using a duplicate protein gene as the bacteria, Spirochaetae, and Posibacteria [M yco­ outgroup for the tree of its sister duplicate is good, plasma genitalium (F ra se r et a l., 1995)], we might there is a practical problem arising from the fact that know the answer by the end of the century, provided the outgroup tree for one duplicate is a very long that whole genomes are sequenced for the other five. branch compared with the branches within the tree It is how;ever possible, as suggested by the 16S rRNA for the other duplicate. It is well known that such a tree, that the radiation was so explosively simul­ very long branch tends to be misplaced on trees taneous that wc shall never be able to resolve the (S w o ffo rd e ta !., 1996). Even if the correct position of order fully, but I am optimistic that we can. the root w'ere at the base of the negibacteria, tree- The fact that if one of these other ideas is correct calculation algorithms will tend to misplace the root and I am w'rong about the rooting of the tree, more nearer to the mean position; the more extreme the of the bacterial high-level taxa proposed here differences in branch length the greater will be this become holophyletic, means that new knowledge tendency. The problem is that the gene duplicates so about the rooting can only increase the attractiveness far studied appear to have diverged very stronglv of the present system to strict H ennigian cladists, not from each other very soon after the duplication and decrease it. However, if Archaebacteria are ever prior to the slower divergence of the major taxa. This shown to be paraphyletic (Lake, 1988, 1989; necessarily means that the outgroup branches are Baldauf, Palmer & Doolittle, 1996), contrary to the 226 T. Cavalier-Smith arguments of both Woese (e.g. 1987) and myself had generally been postulated before (Margulis, (e.g. 1986è, 1987r), the taxon should not be 1970, 1981). Thus primitively amitochondrial abandoned, just because of the dogmas of strict eukaryotic cells must once have existed (Cavalier- Hennigi ans. Smith, 19836). Although there was no guarantee that they had not all gone extinct after the origin of mitochondria, I considered it rather unlikely that IV. PROTOZOA, THE BASAL EUKARYOTIC this would have occurred, since niches for anaerobic KINGDOM phagotrophs must always have existed somewhere in the environment, in which amitochondrial eukary­ Treating Archezoa as a subkingdom of Protozoa otes would have been at no disadvantage compared makes the present kingdom Protozoa very similar in with eukaryotes with mitochondria. Therefore, I composition to Owen’s (1858) original kingdom postulated that at least some of the known amito­ Protozoa, apart from now being properly chondrial eukaryotes were primitively so, and de­ placed with other ochrophyte algae in the kingdom fined the subkingdom Archezoa as comprising all Chromista, and Bacteria being separated into their primitively amitochondrial eukaryotes (Cavalier- own kingdom. It maintains in a single kingdom the Smith, 19836). I was perfectly well aware that some vast majority of the organisms that have been amitochondrial eukaryotes had secondarily lost treated as protozoa over most of the 150 years since mitochondria (i.e. certain and fungi) and Von Siebold (1845) restricted the term to unicellular that some, or conceivably even all, of the taxa that I organisms, as summarized in Table 2. For ap­ initially placed in Archezoa might well also have lost proaching two centuries, since its invention by mitochondria. But in a Popperian spirit I deliber­ Goldfuss (1817), the vernacular term protozoa has ately chose to frame, w hat I explicitly referred to as been so widely used by biologist and layman alike my ‘taxonomic hypothesis', in the most strong and [for example, by the poet Coleridge (1834)], that therefore most easily refutable form by including in there is real value in keeping as close as possible to Archezoa all amitochondrial taxa which then had no the historically dominant meaning without com­ known mitochondrial ancestry. promising the principles of Darwinian classification. Protozoa as a major taxon has, in fact, proved to be (a) Changes m circumscription o f the Archezoa more stable than the remarkable of different interpretations we have seen over the past 50 years of Later, after coming to the view that all hydro- the far more heterogeneous Protista of Haeckel genosomts had probably evolved from either peroxi­ (1866). Protists in Haeckel's sense of unicellular somes or mitochondria (Cavalier-Smith, 1987c), I organisms are now distributed in the present system decided that the double-membraned hydrogeno- across all six kingdoms. However, as I stressed somal envelope of Parabasala was therefore probably previously (Cavalier-Smith, 1981a), the vernacular homologous with the mitochondrial envelope, and word protist remains a most useful term for desig­ therefore removed Parabasala from Archezoa nating the polyphyletic grade of unicellular eukary­ (Cavalier-Smith, 1987a). This change in circum­ otes, even though a kingdom Protista or Protoctista scription of the group, contrary to what is sometimes (in addition to being polyphyletic) would be much implied, did not change the phylogenetic definition too heterogeneous to be taxonomically meaningful. of the group. It was simply a change in view as to whether Parabasala satisfied that definition or not. As previously discussed (Cavalier-Smith, 1993a), the (1) Status of Archezoa, early diverging evidence for a mitochondrial origin for parabasalan amitochondrial eukaryotes hydrogenosomes was, for a long time, rather equivo­ The original concept of the Archezoa was phylo­ cal. However, there Is no good evidence against it, genetic. It was argued that a symbiotic origin of and what is known about protein-targeting to mitochondria could not have occurred until after the hydrogenosomes is consistent with the view that evolution of phagocytosis (Cavalier-Smith, 1983 é). their targeting system evolved from that of mito­ Since phagocytosis is restricted to eukaryotes chondria. This is simpler than the alternative idea of (Stanier, 1961, 1970), it followed that the origin of an entirely independent symbiogenetic event the eukaryote cell and phagocytosis had almost (M üller, 1992), or an even less likely origin from the certainly preceded the symbiotic origin of mito­ endomembranc system. Recent evidence from the chondria (Cavalier-Smith, 19836), not the reverse as proteobacteria! affinities of the heat shock protein A revised six-kingdom system o f life 227 hsp 60 of trichomonads (Bui, Bradley & Johnson et amoebae have each recently been ranked as sub­ a l., 1996; Germot, Philippe & Le Guyader, 1996; phyla within the revised phylum Amoebozoa H o rn e r et a l., 1996; Roger, Clark & Doolitttle, 1996) (Cavalier-Smith, 1997 a ) . In addition to the positive provides the first strong molecular evidence that cladistic evidence for a mitochondrial ancestor, there Parabasala are indeed secondarily amitochondrial, is even more direct evidence from hsp 70 sequences and therefore gives indirect support for a mito­ th a t histolytica evolved from an ancestor chondrial ancestry for their hydrogenosomes. with mitochondria (Clark & Roger, 1995), but in To those who believe that classifications should contrast to the situation in Parabasala these were not not be based on hypotheses, I simply say that all converted into hydrogenosomes. phylogenetic classifications are based on similar sorts Unlike Trichozoa, the Metamonada and Micro­ of phylogenetic hypotheses. The degree to which a sporidia have no hydrogenosomes and branch at the phylogenetic hypothesis needs to be corroborated very base of the eukaryotic rRNA tree, and it has before being used as a partial basis for a classification been widely thought that both taxa are primitively is a matter for scientific judgement by individual amitochondrial. However, I stressed previously systematists. It is neither philosophically correct nor (Cavalier-Smith, 1993a; see also Cavalier-Smith & good manners to call the phylogenetic hypotheses of Chao, 1996a), that this conclusion might not be others speculation, and to treat one’s own as correct. The first evidence, apart from their obligate accepted facts. It is an error to assert that Archezoa intracellular , that Microsporidia might were compositionally defined (Patterson, 1988). be less primitive than was the discovery Although many rRNA trees show Parabasala as of spliceosomal RNA in microsporidia (preliminary branching just below all mitochondrial eukaryotes, data cited in Cavalier-Smith, 1993a; DiM aria et a i , some show them instead as branching just above the 1996) ; this makes it possible that M etamonada are mitochondria-bearing Percolozoa (Cavalier-Smith, the only eukaryotes that primitively lack spliceo­ 1993a,). The latter position is inconsistent with the somal introns. If spliceosomal introns arose from idea that Parabasala are primitively amitochondrial group II self-splicing introns that entered eukaryotes archezoa (that is, if mitochondria are monophyletic), within the genes of the a-proteobacterium ancestral but both positions are consistent with the view that to mitochondria, as postulated (Cavalier-Smith, they are secondarily amitochondrial; which position 1991 c), this would imply that microsporidia evolved is correct is uncertain (Cavalier-Smith & Chao, from ancestors that had mitochondria as previously 1996a), but as several protein trees place the emphasized (Cavalier-Smith, 1993a). Two lines of divergence of Parabasala lower than that of the evidence now strongly support this view and suggest Percolozoa, this position is more likely to be correct. that Microsporidia are really degenerate fungi, and Parabasala are now' treated as a subphylum within must be removed from the Archezoa. First were trees the protozoan phylum Trichozoa, all of which have based on sequences of a-tubulin (Keeling & hydrogenosomes instead of mitochondria (Cavalier- Doolittle, 1996; Li e ta l., 1996) and ^-tubulin (Edlind Smith, 1997a). Trichozoa, Percolozoa and Eugleno­ et a l., 1996; L i et a l., 1996; Roger, 1996), w'hich zoa were recently grouped together as the protozoan group the Microsporidia very robustly with the subkingdom Eozoa to contrast them with the Fungi, not the Protozoa. Second, sequence trees for putatively more advanced protozoa placed in the the molecular chaperone hsp70 equally convincingly subkingdom Neozoa (Cavalier-Smith, 1997a). group the Microsporidia with the Fungi (Germot et The phylogenetic evidence from rRNA that the al., 1997); since the chaperone in question is the archamoebae E ntam oeba, Phreatamoeba (H in k le et al., mitochondrial type related to that of the a-proteo- 1994), and P elom yxa (M orin & M ignot, 1996) are all bacteria, there seems little doubt that microsporidia secondarily amitochondrial made it desirable to are indeed secondarily amitochondrial. A third, but remove the phylum Archamoebae as a whole from much less convincing, piece of evidence is that the the Archezoa, and to place them in the Neozoa protein synthesis elongation factor protein EF 1-a of alongside the amoebae with mitochondria (Lobosa microsporidia has an insertion at exactly the same and Filosea) (Cavalier-Smith, 1997a). The site as does that of all Fungi, animals and choano- evidence that they are secondarily without mito­ zoan protozoa (collectively known as opisthokonta) ; chondria is relatively clearcut, as they branch w'ell however, this insertion is a little shorter and only above the deepest branching point of mitochondrial weakly similar in sequence to that in opisthokonta. taxa. As they sometimes even group with the Lobosa Moreover the EF 1 -a protein sequence tree does not in maximum-likelihood trees, Lobosa and Arch­ group the microsporidia with the Fungi but places 228 T. Cavalier-Smith

them near the bottom of the eukaryotic tree, just as globular and cytoplasmic, except in Uredomycetidae does the small subunit rRNA tree. where they are a unique trilaminar plaque, whereas Thus there is a dramatic conflict between the in Archemycota they are very varied in structure: tubulin and hsp70 trees on the one hand and the EF centrosomal plaques occur in a few Zygomycotina. 1-a and small subunit rRNA trees on the other. The similarity of mitotic mechanism between asco­ Since the two latter trees do not specifically group mycetes and microsporidia was earlier assumed to be microsporidia with any other taxa, and in both cases convergent, partly because a similar mechanism is the microsporidial branches are very long, the also found in the m alaria parasite, , w hich simplest way to reconcile the data are to suggest that must have evolved it independently from Fungi. The the rRNA and elongation factor genes of micro­ mycetozoan Dictyostelium also has centrosomal sporidia have undergone a drastically increased rate plaques, though mitosis is semi-open, so it is clear of evolution compared with all other eukaryotes and that centrosomal plaques have evolved at least three that the deep position in which they appear is a times following independent losses of centrioles. grossly misleading systematic error produced by the Though the mitotic mechanism is clearly, therefore, phylogenetic algorithms, none of which can give the on its own not good evidence for a relationship with correct tree if branches of sisters differ many-fold in ascomycetes and Zygomycotina, it does in con­ length. There are a num ber of other instances where junction with the tubulin and hsp70 sequences rRNA and certain protein trees have been grossly render such a relationship highly probable. misleading, and tree-calculation algorithms have Microsporidia commonly have a binudeate phase been proven to give the wrong answer quite in the life cycle, similar to the dikaryophase of reproducibly in certain circumstances (Felsenstein, ascomycetes and basidiomycetes (Canning, 1990). 1978; Hillis, Huelsenbeck, & Cunningham, 1994; Because in microsporidia the two nuclei are physi­ S w o ffo rd et a l., 1996). cally attached via the surfaces of their nuclear Though initially a surprise, the fungal nature of envelope this binucleate condition is called diplo- microsporidia is not at all unreasonable. Both groups karyotic (Canning, 1990). Some protozoologists have spores with walls ofchitin, and a fair num ber of have suggested that this diplokaryotic condition is a fungi are parasites of animals. A further similarity modification of the dikaryotic condition of higher with higher Fungi is that the vegetative cells do not Fungi, and that microsporidia may be derived from have Golgi dictyosomes visible in the electron higherFungi (Desportes & Nashed, 1983). However, microscope as stacks of cisternae. Higher Fungi are in my view diplokaryosis is distinct from and unusual among higher eukaryotes in lacking Golgi convergent with the fungal dikaryophase (Cavalier- stacks; the only Fungi that have Golgi stacks like Smith, 1995a). most other eukaryotes are the Dictyomycotina, The absence of mitochondria and peroxisomes which include Chytridiomycetes sensu stricto, th e from microsporidia is no reason for excluding them most basal group (Cavalier-Smith, 1987 ¿). The from the Fungi. Both organelles are also absent in absence of Golgi stacks in microsporidia suggests the rumen fungi of the order Neocallimastigales ; that they are not a sister group to the fungi as shown however, these anaerobic fungi do have another by some of the protein trees but, as Roger (1996) respiratory organelle, the , which suggests, actually evolved from within the fungi after may be evolutionarily derived from mitochondria or the allomycete ancestor underwent an evolutionary possibly peroxisomes (Cavalier-Smith, 1987d; unstacking of its Golgi cisternae (Cavalier-Smith, Marvin-Sikkema et al., 1993). Because of the pres­ 1987 ence of Golgi stacks, centrioles and cilia in Neocalli­ It has long been known that the mitotic mech­ mastigales it is most likely that microsporidia lost anism of microsporidia is remarkably similar to that mitochondria and peroxisomes independently of of ascomycetes (Raikov, 1982). In fact their mitosis them, and had either a zygomycote or a hemi- is indistinguishable from that of ascomycetes; in ascomycete ancestor (see section VI). The extremely both groups mitosis is closed with an intranuclear degenerate character of the microsporidian fungi, spindle and a flattened centrosomal plaque at the which caused them to be misclassified for over a spindle poles, either embedded in or just outside the century as protozoa, is one of the most striking nuclear envelope. Most other fungi do not have this examples known of evolutionary degeneration, the type of centrosomal plaque; in basidiomycetes the importance of which was first strongly argued by centrosomes (the recently fashionable fungal term Dohrn (1875). Such degeneration greatly compli­ 'spindle pole body’ is quite unnecessary) are cates the task of phylogenetic reconstruction, as A revised six-kingdom system o f life 22 9

Lankester (1877, 1880) emphasized, since its at­ heat shock protein ; a-tubulin, /9-tubulinl. At present tendant simplification and loss of ancestral charac­ there is no evidence that they have any spliceosomal ters is so easily confused with primitive simplicity. introns, unlike all higher eukaryote phyla. Time and again secondarily simplified organisms Given these distinctive features and their ap­ have been misclassified with genuinely primitive parently early divergence I here revise the sub­ ones. kingdom Archezoa to include both Metamonada The clear demonstration that Trichozoa, Arch­ and Trichozoa (as it did originally: but now amoebae and Microsporidia are all secondarily Microsporidia and Archamoebae are excluded). amitochondrial leaves only the phylum Meta­ Though it now appears that Archezoa in this revised monada, which appears to have diverged from all sense are secondarily without mitochondria, there is other eukaryotes earlier than any others (Cavalier- no evidence that they ever had peroxisomes. It is Smith & Chao, 1996 ö ), as possibly primarily possible therefore that they are primarily without amitochondrial. Transferring Microsporidia to the peroxisomes, unlike all higher eukaryotes. If this is kingdom Fungi, left only Metamonada in the true and if, as De Duve (1984) and I (Cavalier- Archezoa, which I now treat as a subkingdom of Smith, 1987c) have argued, peroxisomes are derived Protozoa (Cavalier-Smith, 19976), as it was orig­ symbiogenetically from bacteria, then Archezoa may inally (Cavalier-Smith, 1983a), rather than as a be less chimaeric in origin than all higher eukaryotes. separate kingdom (Cavalier-Smith, 1987a; 1993a). In postulating that the presently redefined Arch­ If M etamonada lack both mitochondria and peroxi­ ezoa may be primitively without spliceosomal introns somes primitively (Cavalier-Smith, 1983a), then and peroxisomes, I do not rule out the possibility they would be radically different from all higher that they have lost them, or that some spliceosomal eukaryotes (collectively designated M etakaryota: introns may one day be found. I am, as in my earlier Cavalier-Smith, 1987a). But it seems increasingly discussions of the Archezoa, merely following Lankes­ unlikely that they are indeed genuinely primitively teri (1877) methodologically sound principle ‘that amitochondrial, contrary to widespread assump­ until we have special reason to take a different view of tions. The published evidence that any particular case we are bound to make the may have arisen by the loss of mitochondria (Keeling smallest amount of assumption by assigning to the & Doolittle, 1987; Soltys & Gupta, 1994) is not yet various groups of organisms the places which they very convincing, but firmer sequence evidence for a will fit into, on the supposition that they do represent secondarily amitochondrial character of the group in reality the original progressive series’. may soon be forthcoming from heat shock proteins (Roger and Sogin, personal communication) and aminoacyl iRNA synthetases (Hasegawa, personal (2) The protozoan subkingdoms: Archezoa communication). a n d N e o z o a If, as seems likely, Metamonada are indeed secondarily amitochondrial, then the distinction The recent grouping of the phyla Trichozoa, between them and the Parabasala becomes less Euglenozoa and Percolozoa together as the sub­ fundamental than I thought when I removed kingdom Eozoa (Cavalier-Smith, 1997 a) was heavily Parabasala from the Archezoa on the grounds that influenced by the fact that on rRNA trees these three their hydrogenosomes probably arose from mito­ taxa consistently diverge earlier from all other chondria (which is now generally accepted). M ore­ mitochondrial eukaryotes (neokaryotes: Cavalier- over, Archezoa can no longer be distinguished from Smith, 19936) than the latter do from each other. other eukaryotes on the grounds that they are However the recent demonstration that Micro­ primitively amitochondrial: we must therefore re­ sporidia are Fungi not Archezoa, shows just how consider the definition and circumscription of the grossly misleading the rRNA tree can sometimes be. Archezoa and the other protozoan subkingdoms This highlights eariy arguments that rRNA gene (Cavalier-Smith, 1997a). Metamonada and Tri­ evolution must be substantially non-clock like and chozoa are anaerobic or microaerophilic flagellate subject to gross systematic biases in evolutionary rate protozoa that lack mitochondria and have kinetids (Cavalier-Smith, 1980). The recent demonstration that ancestrally have four centrioles, not two as in that the histionid flagellate americana h as higher eukaryotes. They also appear to be the two the most primitive known mitochondrial genome earliest branching eukarvote phyla on several dif­ organisation (Lang et a t., 1997), when coupled with ferent protein sequence trees (e.g. RNA polymerase ; the fact that rRNA trees place Reclinomonas a b o v e 230 T. Cavalier-Smith both Percolozoa or Euglenozoa (Cavalier-Smith & diverging eukaryote phyla are the M etamonada and Chao, unpublished) adds further fuel to my earlier Trichozoa. To retain separate subkingdoms for these suspicions that rRNA trees may place the Euglen­ two taxa (Archezoa and Eozoa) seems less useful ozoa substantially too low (Cavalier-Smith, 1993c, than to group them together in a single subkingdom, 1 9 9 5 a ). R u lim m o n a s is the only known eukaryote to as suggested above. Archezoa thus remains as the have retained the a-proteobacterial RNA poly­ basal paraphyletic protozoan and eukaryote sub­ merase genes in its mitochondrial genome. It seems kingdom, as in the original 6-kingdom system unlikely that their replacement by nuclearly encoded (Cavalier-Smith, 1983a) ; but its phylogenetic defin­ phage T3/T7 type RNA polymerases took place ition is modified - Archezoa comprise the two phyla more than once in other eukaryotes. This makes it that diverged prior to the divergence of all the likely that Euglenozoa and Percolozoa arose from mitochondria-containing groups. Contrary to what neozoan ancestors in which this had already oc­ was previously suggested (Cavalier-Smith, 1983ö), curred. If true, then the rRNA tree is grossly the latest common ancestor of the Archezoa had misleading as to their position and the Eozoa are probably already started to incorporate at least some polyphyletic rather than paraphyletic as earlier of the genes from the symbiont that was eventually- suggested (Cavalier-Smith, 1997 a). In view of this, I converted into the , and therefore was here remove the Euglenozoa and Percolozoa from probably not a non-chimaeric eukaryote, as postu­ the Eozoa and group them together as a new lated earlier (Cavalier-Smith, 1983¿). W hether this infrakingdom, Discicristata, which I place in the archezoan cenancestor had a fully developed mito­ subkingdom Neozoa. chondrion or only a precursor of mitochondria is a Such a position is consistent with my earlier semandc rather than a factual issue. In my view argument that the triple enveloped chioroplast however it is proper to call the common ancestor of envelope of dinoflageliates and euglenoids is a unique the trichomonad hydrogenosome and the mitochon­ shared derived character that links the two groups, drion a mitochondrion since it must have been at and which could have arisen through a primary least facultatively aerobic with cytochromes and endosymbiosis, not a secondary endosymbiosis as must already have evolved an organelle-specific commonly assumed (Cavalier-Smith, 1982, 1995a). protein import mechanism of the type shared The presence of articulins in both (pre­ between the two organelles (Bradley el a t., 1997), the sently known only in ciliates) and euglenoids is best demarcation criterion between an obligate consistent with a cioser relationship between Alve­ symbiont and an organelle. Therefore trichomonad olata and Discicristata than is suggested by the 18S hydrogenosomes did evolve from mitochondria, as I rRNA tree. The fact that certain protein trees, postulated (Cavalier-Smith, 1987if). The suggestion notably tubulins and EF 1-a (Baldauf & Palmer, that the accepted common ancestor of these hydro­ 1993) place Euglenozoa and Percolozoa as a single genosomes and mitochondrion might not be a clade near the green plants and not below the mitochondrion but merely a precursor of a mito­ neozoan radiation, is consistent with a primary chondrion (Bui, Bradley & Johnson, 1996) is a origin for the euglenoid chioroplast (Cavalier-Smith, distinction without substance since no properties are 1982) and favours the view that 18S rRNA trees suggested by which the ‘non-mitochondrial’ pre­ place Discicristata too low, just as they do Micro­ cursor might be distinguished from a mitochondrion. sporidia and Mycetozoa. Though the tetrakont Even though one cannot (at least at present) deduce character of some Percolozoa and the absence of whether or not the mitochondrial protein-import Golgi dictyosomes were reasons for suggesting earlier mechanism had evolved prior to the divergence of that they might be the most primitive mitochondrial M etamonada and Trichozoa, it is probably best not eukaryotes (Cavalier-Smith, 1993c), the recent mito­ to continue to think of Archezoa as being primitively c h o n d ria l d a ta fo r Reclinomonas suggest that histionids amitochondrial. However, the reasons for postu­ may be more primitive. Detailed studies are needed lating that the a-proteobacterium was taken up by of mitochondrial genomes and a variety of nuclear an early amitochondrial and non-chimaeric eukary­ protein-coding genes in Percolozoa and a much ote after the origin of the cytoskeleton, endo- greater variety of tubulicristate flagellates in order to membrane system and phagocytosis (Cavalier- test this and the present classification more Smith, 19836, 1987(7) rem ain compelling, even if no thoroughly. primitively amitochondrial descendants of this early If Discicristata are really part of the neozoan non-chimaeric phase of eukaryote evolution remain. radiation, as argued here, then the only really early IfDiscicristata really occupy the position shown in A revised six-kingdom system o f life 231

Fig. 1, then it is necessary to redefine the term ata and Bilateria and ranked as a third subkingdom neokaryotes, originally defined as the clade including of the kingdom Animalia (Cavalier-Smith et a i , all eukaryotes that branch higher on rRNA trees 1996 ri). The great difference in phenotype between than Euglenozoa {Cavalier-Smith, 1993Í}, because Myxozoa and both Radiata and Bilateria also this definition no longer defines a clade. I suggest justifies this high rank; to place them in either that neokaryote should now signify the clade subkingdom would make it phenotypically too comprising Neozoa plus the four higher kingdoms of heterogeneous. Long ago Lankester (1877) noted life. Eukaryotes may therefor be divided simply into that it was very hard to disprove the idea ‘ that many tetrakont Archezoa, putatively without spliceosomal of the Protozoa are not descended from Enterozoa introns, and the basically bikont neokaryotes, typi­ by degeneration ’ : it appears that only the Myxozoa cally with spliceosomal introns (with the exception have actually done so. of the euglenozoan Kinetoplastea, which, probably Recent molecular phylogenetic evidence that lost them. The present redefinition of both Archezoa dicyemid Mesozoa (Katayama et a i , 1995) and and neokaryote make the terms eokaryote (Cavalier- orthonectid Mesozoa (Hanell et a i , 1996) are related Smith 19936) and megakaryote (Cavalier-Smith, to bilaterians means that they also have lost a 1995a) no longer necessary. M etakaryota (Cavalier- nervous system, gut and collagenous connective Smith, 1987a) are not a taxon in the present system ; tissue, another remarkable example of degeneration but following the recognition that microsporidia and raised as a possibility by Lankester. Though the Archamoebae are secondarily amitochondrial and molecular data suggest that the two mesozoan classes branch higher up molecular sequence trees than do may not be directly related to each other (Hanelt et Parabasala, the name metakaryotes (Cavalier- a i , 1996) the trees lack strong resolution, so the Smith, 1987 a) must now include both microsporidia closest relatives of both groups are uncertain and it and archamoebae and, therefore, now designates the remains possible that Mesozoa are monophvletic; clade comprising all eukaryotes other than M eta­ th e re fo re I retain the phylum Mesozoa until such m o n a d a . time as its monophyly is more convincingly dis­ proved, but in view of the radical differences from Bilateria continue to place it in a distinct sub­ (3) Demarcation between the two zoological kingdom Mesozoa (Cavalier-Smith, 1983ö, Cava­ kingdoms: Protozoa and Animalia lier-Smith et al., 1996a). Whether Mesozoa are Now that the phylogenetic position ofM yxozoa has monophyletic or not, it is now clear that dicyemids, been established by rRNA phylogeny (Smothers et orthonectids and M yxozoa all arose by the loss of the a l., 1994; Schlegel el al. 1996), it is clear that their nervous system and of obvious collagenous tissue; vegetative unicellularity is secondarily derived as a thus it is no longer justifiable to use the absence of result of parasitism. It seems to be only a remote collagenous connective tissue as a reason for ex­ possibility that their position on rRNA trees well cluding Mesozoa from Animalia (Cavalier-Smith, within the metazoan animals is a methodological 1983a, 1993a). The complex life cycles of the artefact of long-branch attraction. While other Mesozoa and their copulatory sex have long been molecular data are clearly needed, the present data reasons for considering that they may have evolved are convincing enough to make it highly probable from a -like ancestor by the loss of the that Myxosporidia arose from a multicellular an­ nervous system (Stunkard, 1954; 1972); unfortu­ cestor that possessed not only cell junctions, as do nately the base of the bilaterian rRNA tree is a bush­ their multicellular spores, but also collagenous like radiation in which no branching orders are connective tissue, a nervous system and a gut. clearly resolved, which is consistent with the explo- Therefore Myxozoa must be excluded from the sivencss of the early Cam brian radiation seen in the kingdom Protozoa and placed within the kingdom fossil record. The fact that their mitochondrial Animalia, as stated earlier (Cavalier-Smith, crisiae are tubular not flat, evidence previously cited 1 9 9 5 ¿>, c). Since it is unclear whether they are in support of a protozoan character (Cavalier-Smith, derived from Cnidaria, as is suggested by their 1983a), is not sufficient reason for excluding them nematocyst-like extrusomes (Weili, 1935; Lorn & De from Animalia, since the crisiae of can Puytorac, 1965; Siddall et a i , 1995) or are closer to also be tubular rather than the flat cristae of most bilateral animals, as is weakly suggested by the animals other than Ctenophora. rRNA trees, at present Myxozoa are most ap­ The generalization that animals have flat cristae propriately excluded from both subkingdoms Radi­ (Taylor, 1978) is clearly an oversimplification. It has 232 T. Cavalier-Smith

Table 3. Classification o f the kingdom Protozoa] and its 13 phyla

Subkingdom 1. ArchezoatCavalier-Smith 1983 cm. Phylum 1. M etamonada Grassé 1952 slat. nov. ct cm. Cavalier-Smith 1981. Subphylum 1. Eopharyngia Cavalier-Smith 1993 (e.g. , , Trepomonas, ). Subphvlum 2. Axostylaria Grassé 1952 slat. nov. em. Cavalier-Smith 1993 (e.g. , ). Phylum 2. Trichozoa Cavalier-Smith 1997. Subphylum 1. Anaeromonada Cavalier-Smith 1997 ! ). Subphylum 2. Parabasala Honigberg 1973 slat, nov. Cavalier-Smith 1997 {e.g. , ), Subkingdom2 . N e o z o a t Cavalier-Smith 1993 stat. nov. 1997 em. Infrakingdom 1. Sarcomastigota* Cavalier-Smith 1983 slat. nov. em. [diagnosis: typically sarcodines, flagellates, or amoehoflagellates; usually free-living , rarely chimacric photophagolrophs with nticleoinorphs and periplaslid membranes; pseudopods varied, but seldom eruptive, typically aerobes with tubular, less often flat or rarely discoid, mitochondrial cristae and prominent Golgi dictvosomes: cortical alveoli absent; axopodia usually absent iii present with quincunx arrangem ent;: amitochondrial species lack hydrogcnosomcs and well-developed diciyosomcs]. Phylum 1. Neomonada* Cavalier-Smith 1997. Subphylum 1. Cavalier-Smith 1997 ie.g. , , , Reclinomonas Ebria). Subphvlum 2. Isomita Cavalier-Smith 1997 (e.g. , Cyathobodo, Kathablepharis). Subphylum 3. Choanozoa* Cavalier-Smith 1981 cm. 1983 stat. nov. (c g. Monosiga, Diaphanoeca, Corallochylrium, Psorospermium). Phylum 2. Cercozoa phyl. nov. (syn. Rhizopoda Von Sicbold 1845 stal. nov. Haeckel 1886 as emended by Cavalier-Smith 1 9 9 5b, 1997a; present phylum emended by addition of Spongomonadida and transfer of Cristidiscoidia to Choanozoa: Cavalier-Smith, 19976.) (diagnosis: unicellular phagotrophic heterotrophs or else photosynthctic algae with green chloroplasts and nucleomorphs within a pcriplastid membrane located inside a fourth smooth membrane; typically free-living aerobes having peroxisomes and mitochondria with tubular (or very rarely flat or vesicular) crisiae; flagellates with two usually anisokont cilia or a single cilium or non-flagellates (usually rhizopods) with a lest and/or filóse or reticulosc pscudopodia or with a green plaslid and nucleomorph; cilia without lateral flanges, paraxial rods, transition helix or tubular hairs; cortical alveoli and axopodia absent: heterotrophs have a flexible cell surface without a rigid dense protein layer inside or outside the plasma membrane; distinct cytopharynx absent; silica scales sometimes present but internal silica skeleton absent: extrusomes, if present, isodiammetric or a complex Stachel; often with walled cysts). Subphylum 1. Phvlomyxa Cavalier-Smith 1997 (e.g. ). Subphylum 2- Reticulofilosa Cavalier-Smith 1997 ¡e.g. Chlorarachnion, Cryptochlora). Subphylum 3. Monadofilosa Cavalier-Smith 1997 (e.g. Cercomonas, Gymnophrys, , Spongomonas). Phylum 3. (D'Orbigny 1826) Eichwald 1830 stat. nov. Margulis 1974 (e.g. , Ammonia). Phylum 4. Amoebozoa Ltihc 1913 slat. nov. em. [emended diagnosis solitary or aggregative amoebae with usually non- cruptivc lobose pseudopods, not uncommonly with finer pointed subpseudopodia; mitochondria with tubular crisiae, or sometimes absent: Golgi diciyosomcs well-developed except in Archamoebae; sometimes with transient or permanent cilia; usually one cilium (rarely two) per kinctid and one kinetid (rarely many) per cell]. Subphylum 1. Lobosa Carpenter 1861 stat. nov. Cavalier-Smith 1997 cm. (e.g. , , , , Multicilia). Subphylum 2. Conosa subphyl. nov. (diagnosis: kinctid, when present, with a eone of microtubulcs which typically subtends the nucleus at its broader end, and with a lateral microlubular ribbon closer to the cell surface; aggregative aerobes with mitochondria or solitary anaerobes lacking mitochondria and peroxisomes). Infraphylum 1. Archamoebae Cavalier-Smith 1983 slat, nov (e.g. , , Phreatamoeba, Entamoeba). Infraphylum 2. Mycetozoa De Bary 1859 stat. nov. Superclass 1. Eumyxa* Cavalier-Smith 1993 stat. nov. (e.g. Protostelium, Physarum). Superclass 2. Dictyostclia Lister 1909 stal. nov. (e.g. Dnlyosleliumi Infrakingdom 2. Discicristata infraking. nov. (diagnosis: mitochondrial cristae typically discoid ; pseudopods if present are eruptive lobes; kinetid typically with two centrioles, rarely four or absent). Phylum 1. Percolozoa Cavalier-Smith 1991. Subphylum 1. Tetramitia Cavalier-Smith 1993 (e g , Saegleria). Subphylum 2. Pseudociliata Cavalier-Smith 1993 ( ). Phylum 2 Euglenozoa Cavalier-Smith 1981 Subphylum 1. Plicosloma subphyl. nov. (diagnosis: with pellicular strips and/or a feeding apparatus comprising two or three supporting rods and four or five curved or plicate vanes. (This formalizes the name plicostomc, first used by Patterson ; 1988) for this taxon. even though it is not ideal as the vanes are often not plicate and are absent from pctalomonads) [superclasses Diplonemia Cavalier-Smith 1993 stal. nov. i,pellicular strips absent, e.g. Diplonema) and Euglenoida Butschli 1884 ¡tai. nov. Cavalier-Smith 1997 (pellicular strips present, e.g. , Petalomonas, )]. Subphylum 2. Saccostoma subphyl. nov. (diagnosis: subapical and cylophyarvnx reinforced on one side by microtubulcs (the so-called microlubular root (M TR)/pockel feeding apparuratus: hence the name from latin saccus, bag and stoma mouth), lacking dense reinforcing rods or vanes; without pellicular strips). Classes Kinetoplastea Honigberg 1963 slat, nov. Margulis 1974 (e.g Rodo, , ) and Postgaardea cl. nov. (diagnosis: mitochondria without kinctoplasl or crisiae: sole genus : S im p so n et al., 1997). A revised six-kingdom system o f life 233

Table 3. ( conl .)

Infrakingdom 3 . A lv e o la ta Cavalier-Smilh 1991 cm. Supcrphylum 1. Miozoa Cavalier-Smilh 1987. Phylum 1. Dinozoa Cavalier-Smith 1981 em

Subphvlum 1. Protalveolata* Cavalier-Smith 1991 cm. (C R. , , Spironema, Hemimaslix, , ). Subphylum 2. Dinoflagellata Bülschli 1885 stal nov. Cavalier-Smilh 1991 (e.g. .Yoctiluca, Crypthecodinium, j. Phylum 2. Sporozoa Leuckart 1879 stat. nov. Cavalier-Smilh 1981 :syns Tclosporidia Schaudinn 1900; Euspora Levine 1961; Polannulifera Levine 1969; Levine 1970 pro parte). Subphylum 1. Grcgarinac Haeckel 1866 stal. nov. (e.g. Monocystis). Subphylum 2. Coccidiomorpha Doflein 1901 [Superclasses I.euckart 1879 stat. nov. Cavalier-Smith 1993 {e.g. Sarcocyslis, Toxoplasma, ), Ascetospora Sprague 1979 scat. nov. ie.g. Haplosporidium, Paramyxa) and Hemaiozoa Vivier 1982 stat. nov. ieg. Plasmodium, )]. Subphylum 3. M anubrispora subphyl. nov. (Diagnosis: plasmodia] intracellular parasites of gregarines with inner membrane complex in vegetative cells, but no apical complex, mitochondria, peroxisomes, centrioles or cilia; uninucleate spores usually spherical, with separate membrane-bounded polar body and m anubrium/lamellar complex, a thin dense wall; spores formed within walled cyst by plasmolomy; Golgi complex an aggregate of vesicles, associated with centrosomal plaque during closed mitosis; spindle intranuclear: sole class Mctchnikovellea Weiser 1977 emend. Cavalier-Smilh 1993 to exclude M inisporca, e.g. Metchnikovella). Supcrphylum 2. Heterokaryota Hickson 1903 stat. nov. Cavalier-Smith 1993. Phylum Ciliophora Doflein 1901 stat. nov. Copeland 1956 cm. auct. Subphylum 1. Tubulicorticata de Puytorac el al. 1992 (e.g. , , ). Subphylum 2. Epiplasmata de Puylorac et al. 1992 (eg. , , ). Subphylum 3. Filocorticata de Puytorac et al. 1992 (e.g. Spathidium). Infrakingdom 4. AdinopodaCalkins 1902 stal. nov. Cavalier-Smith 1997. Phylum 1. Haeckel 1886 stal. nov. Margulis 1974 (e.g. Actinophrys, Acanthocystis) Phylum 2. Radiozoa Cavalier-Smith 1987. Subphylum 1. Spasmaria Cavalier-Smith 1993 [, acantharians, e.g. Acanthometra). Subphylum 2. M üller 1858 emend, stat. nov. Cavalier-Smith 1993 (e.g. Thallassicolla, Aulacantha}.

* Probably paraphyletic. t Almost certainly paraphyletic. For a more detailed classification of protozoa to the level of subclass and discussion of its phylogenetic basis sec Cavalier- Smith (1993/1, 1995A, 1997a, b ) and Corliss (1994). The name Archezoa (Cavalier-Smith 1983û) is derived from the Greek arehe meaning ‘beginning’ or ‘first’ (as is archetype), n o t fro m archaias imeaning ancient as in archacologv) ; the spelling Archaezoa sometimes seen ie.g M aynard Smith & Szalhmary, 1995) is incorreci. When I created the taxon I was perfectly aware thai Perly (1852) had used Archezoa in a broader sense; bul as nobody since Haeckel (1868 and subsequent editions) had used it at all, 1 considered that no confusion would arise. In any ease, it is often desirable to refine old names rather than to invent totally new ones; any temporary confus­ ion soon fades - nobody now complains that the name scarcely overlaps at all with Linnaeus's melange grouped under lhal name, or that Insecta, Crustacea and Chordata no longer have the same circumscription as they once did, or that Linnaeus included bats among ihe Primates, and Reptilia and many fish in his Amphibia, whilst his Reptilia included and tortoises but excluded snakes. If we followed those pedantic nomenclalurists who think that all changes in circumscription of higher taxa necessarily require them lo be renamed, we would have to change all ihese and many other established names thus causing undue confusion. As in ihe present svstem, Haeckel defined Archezoa as the most primitive Protozoa: but he thought that amoebae were the most primitive organisms and thus, unlike Cavalier-Smith (1983a). excluded flagellates from Archezoa. It is now clear that all amoebae arc derived from flagellates, contrary to Haeckel’s view, so flagellates are more- primitive than amoebae, and all non flagellates except Dientamoeba are excluded from the present Archezoa.

long been known that those of Ctenophora are flagellates or the tubular cristae of most other distinctly tubular as are those of adrenal neozoan protozoa that it would be more accurate glands. Clearly, tubular cristae have evolved from simply to call them irregular. The origin of the the ancestral animal condition of fiat cristae several irregular cristae of myxozoa from the more typical times within the kingdom Animalia. The cristae of flat cristae of most low'er invertebrates might have myxosporidia are sometimes referred to as flat been contemporaneous with the secondary origin of (Corliss, 1984) and sometimes as tubular (Taylor, their amoeboid plasmodial vegetative state ; ac­ 1978), but they are usually not clearly either. They cording to the molecular coevolutionary theory of are so different from the flat cristae of choano- the origin of mitochondrial cristae, a marked change 234 T. Cavalier-Smith in cristal morphology is often a pleiotropic neutral Botanical Code of Nomenclature (Cavalier-Smith, consequence of adaptive changes in the ceil surface 1981a). (Cavalier-Smith, 1997a). The inclusion of Archezoa and the exclusion of (4) Classification of the Neozoa Myxozoa and Mesozoa from the kingdom Protozoa (al The infrakingdom Sarcomastigota means that the k in g d o m now includes all primitively unicellular phagotrophic non-photosynthetic Neozoan classification is still in a state of flux. eukaryotes that have not evolved by the secondary Ribosomal RNA sequence studies on a wide variety loss of chloroplasts, and is predom inantly m ade up of of zooflagellates currently in progress in our lab­ such organisms, i t is thus a relatively homogeneous oratory indicate that the distinction between the grade of organisation that will continue to be a very infrakingdoms Sarcodina and Neomonada is less useful m ajor unit of classification. The fact that it is clear than it seemed earlier (Cavalier-Smith, 1997a). obviously paraphyletic, in contrast to the holo- It therefore seems sensible to merge them into a phyletic kingdom Animalia, does not lessen its utility. single infrakingdom, for which I adopt the name The division of the phagotrophic zoological world Sarcomastigota, used earlier (Cavalier-Smith, into two distinct kingdoms, the unicellular Protozoa 1983 a) for a similar assemblage of flagellates and and the ancestrally triploblastic Animalia, appro­ sarcodines. Unlike my earlier subkingdom Sarco­ priately recognizes the fundamental differences in mastigota (Cavalier-Smith, 1983a), which was not body plan between the two zoological kingdoms. generally adopted, the present infrakingdom Sarco­ The higher level classification of each is summarized mastigota includes Choanozoa since the distinction in Tables 3 and 4. W hether one adopts some variant between tubular and flat cristae is less fundamental of the five-kingdom system (W hittaker, 1969} or the than was earlier thought (for the reasons, see six-kingdom system (Cavalier-Smith, 1981 a, 1983a), Cavalier-Smith, 1997a); it also excludes the Alveo­ the exclusion of Protozoa from the kingdom Ani­ lata and Adinopoda, which are treated as separate malia means that Haeckel's term M etazoa is now an infrakingdoms. Here I follow Siddall, Stokes & unnecessary synonym of Animalia. It is desirable to Burresson (1995) in placing Haplosporidia in the harmonize the circumscription of the vernacular Alveolata; even though this assignment is most term animal with that of the formal term Animalia uncertain (Cavalier-Smith, 1997a), excluding them and not to include protozoans within its ambit. Even from the infrakingdom Sarcomastigota makes it in the most elem entary teaching, it is best not to refer phenotypically more uniform. to protozoa as unicellular animals but to encourage In the present system Mycetozoa are no longer the view of protozoa as a different and more treated as a separate phylum; instead they are primitive form of life than animals, plants or fungi. placed within the phylum Amoebozoa, and grouped Calling protozoa unicellular animals not only down­ with the Archamoebae as a new subphylum Conosa, plays the great significance of the differences in body characterized in the flagellate members of the g r o u p plan between the two kingdoms, but helps to by a eone of microtubules emanating from the often condemn the study of protozoa to a position of single centriole and subtending the nucleus. It secondary importance within zoology, which their appears that rRNA trees usually place the Conosa status as a separate kingdom belies. Z o o lo g y is th e m u c h to o lo w ; protein trees place them much higher study of animals and protozoa, each of which have up near the opisthokonta (animals, fungi, Choano­ important but distinct roles in the biosphere. The zoa) and in the case of actin group them together. phrase ‘unicellular animals’ is also particularly Only two subphyla are recognised in the Amoebo­ inappropriate for the three groups of protozoan zoa: Conosa and Lobosa. The Lobosa comprise the algae: Euglenia, Chlorarachnea and photosynthetic three classes Amoebaea, Testacealobosea and Holo- dinoflagellates. For the reasons discussed previously mastigea (with sole genus M u ltic ilia, which is now (Cavalier-Smith, 1993 a, 1995 a), these three photo­ placed in the Lobosa as it has cell surface glycostyles synthetic groups are too closeJv related to non- similar to those of some Amoebaea: Mikryukov & photosynthetic protozoa to be excluded from the Mylnikov, 1996a ). Following these simplifications kingdom. Since the term algae has long since ceased there are now only four sarcomastigota phyla: two to refer to a botanical taxon, there is no contradiction essentially sarcodine (Foraminifera, Amoebozoa) in referring to these three groups as protozoan algae and two predominantly flagellate but with a signifi­ or in suggesting that, like all other Protozoa, they cant sprinkling of rhizopods or other non-flagellates should be subject to the Zoological, not to the (Neomonada and Cercozoa). A revised six-kingdom system o f life 235

Table 4. Classification of the kingdom Animalia and its 23 phyla

Subkingdom 1. R a d ia t a j Linnaeus 1758 em. siat. nov. Cavalier-Smilh 1983 (multicllellular; radial or biradial symmetry; no anus). Infrakingdom 1. Spongiaria*De Blainville 1816. Phylum Porifera Grant 1836 (sponges). Subphylum 1. Hyalospongiae Vosmaer 1886 slat. nov. cm. [typically with silicious spicules; sometimes also calcified (i.e. pharetronids; sphinctozoans are probably secondarily without spicules) but without calcareous spicules; mainly dcmosponges, e.g. A xinella, and hexactinellids]. Subphylum 2. Calcispongiae Blainville 1834 slat. nov. (syn. Calcarea Bowerbank 1864 e.g. Clathrina, Sycon). Subphylum 3. Archacocyatha Vologdin 1937 (extinct). Infrakingdom 2 . Coelenterata* Leuckart 1847 em. auct. Phylum 1. Cnidaria* (corals, sea anemones; jellyfish, hydrolds). Subphylum 1. Anthozoa* Ehrcnbcrg 1831 stal. nov. (e.g. Sarcophyton. Actinia). Subphvlum 2. Medusozoa Petersen 1979 (e.g. Hydra, Physalia, Tripedalia, Aurelia). Phylum 2. Ctenophora Eschscholtz 1829 (comb jellies, e.g. Beroe, Pleurobrachia). Infrakingdom 3 . P la c o z o a infrak. nov. (without body cavity, gut or nervous system). Phylum Placozoa Grell 1971 ( ). Subkingdom 2. MyxozoaG ra ss e 1970slat. nov. Cavalier-Smilh 1996 (unicellular non-ciliatc parasites with m ulticellular spores). Phylum M yxosporidia Bütschli 1881 slat. nov. Grasse 1970 (e.g.M yxid iu m ). Subkingdom3 . Bilateria* Hatschck 1888 stal. nov. Cavalier-Smith 1983 (bilateral animals, primitively with an anus and probably coelom). Branch 1. Protostom ia*Grobben 1908. Infrakingdom 1. L o p b o z o a *infrak. nov. (diagnosis: primitively sessile with U-shaped gut and ciliated oral tentacles with coelomic extensions; early ciliated larvae trochophorcs, later often bivalved). Supcrphylum 1. Polyzoa Thompson 1830 stat. nov. cm. (diagnosis: no vascular system or longitudinal nerve cords; adult without shell; zooids able to multiply by vegetative budding, often colonialis", larval brain disappears at metamorphosis). Phylum 1. * Ehrenberg 1831 (unnecessary junior synonym: Ectoprocta Nitsche 1870). Subphvlum 1. Gymnolaemata Allman 1856 stat. nov. (e.g. Bugula, Flustra, Membranipora). Subphvlum 2. Lophopoda Dumortier 1835 (syn. Phylactolaemata Allman 1856, e g- Plumatella). Phylum 2. Kamptozoa Corl 1929. Subphylum 1. * Nitsche 1870 (e.g. Loxosoma, Pedicellina, Urnatella). Subphvlum 2. Cycliophora Funch & Kristcnsen 1995 slat. nov. {). Supcrphylum 2. Conchozoa supcrphyl. nov. (diagnosis: vascular system; ancestrally with calcareous shell, primitively bivalved and unhinged). Phylum 1. Mollusca Linnaeus 1758 em. Lamarck. Subphylum 1. Bivalvia* Linnaeus 1758 stat. nov. em. auct. (bivalves, e.g. Mytilus, Pecten: unneccessarv junior synonyms Acephala Cuvier; Lipocephala Lankester 1889). Subphvlum 2. Glossophora Lankester 1889 stat. nov. (with radula and head). Infraphylum 1. Univalvia Linnaeus 1758 stat. nov. em. (diagnosis: non-chambcrcd shell in one piece; tentacles; crystalline style: monoplacophorans, gastropods, scaphopods). Infraphylum 2. Spiculata infraphyl. nov. (diagnosis: calcareous spicules and/or multiple shell plates; without eyes, tentacles or crystalline style: aplacophorans, caudofoveatus, polyplacophorans). Infraphylum 3. Cephalopoda Cuvier 1797 (single multichambercd shell, e.g. octopus, squid, nautilus, cuttlefish). Phylum 2. Brachiozoa phyl. nov. (diagnosis: with lophophorc and vascular system). Subphylum 1. Brachiopoda* Dumeril 1806 (e.g. Lingula, Terebralula). Subphylum 2. Phoronida Halschek 1888 ( Phoronis, Phoronopsis). Supcrphylum 3. supcrphyl. nov. (U-shaped gut, ventral non-gangllonated nerve cord; no vascular system or shell; pelagosphaera larva). Phylum Sipuncula Raffincsque 1814 slat. nov. Sedgwick ¡898 (e.g. Golfingia, Phascolosoma). Supcrphylum 4. Vermizoa superphyl. nov. (diagnosis: coelomatc worms with blood and straight gui with anus; riliated larvae without bivalved shells; two ventrolateral or one primitively paired ventral nerve cord).

[continued overteaj 236 T. Cavalier-Smith

Table 4. (cont.)

Phylum I. Annelida Lamarck 1809 (unnecessary junior synonym Chaclopoda: ancestrally segmented with coelom around gut; chaetac; ganglionated ventral nerve cord). Subphylum 1. Polychaeta* Grube 1850. Infraphylum 1. Operculata infraphyl. nov. (diagnosis: live in calcareous lubes with operculum; lacking muscular pharynx, e.g. Spirorbis, Serpula j. Infraphylum 2. Pharyngata infraphyl. nov. (diagnosis: with muscular pharynx; if tube-dwellers without calcareous operculum: bristlcworms, e.g. Nereis, Arenicola , Sabella) . Subphylum 2. Clitellata auct. (earthworms, leeches). Subphylum 3. Echiura Sedgwick 1898 orth. cm. Stephen 1965 stat. nov. (e.g. Urechis, Bonellia ). Subphylum 4. Pogonophora Johannson 1937 stat. nov. (Frenulata and Vestimentifera, e.g. R iftia ). Phylum 2. Nemertina Oersted 1844 em. Schullzc 1850 (unnecessary syn. Rhynchocoela Schulze 1850) (unsegmented coclom around only cversiblc proboscis: proboscis worms, e.g. Nem ertes). Infrakingdom 2 . Chaetognathi Leuckart 1854 stat. nov. Phylum Hyman 1959 (arrowworms, e.g. Sagitta, Spadella). Infrakingdom 3 . E c d y s o z o ainfrak. nov. (diagnosis: with thick cuticle that is moulted; no surface cilia or ciliated larvae; gut, if present, with anus; coelom present only embryonically or absent; ventral nerve cord; usually gonochoristic). Name suggested for this clade minus by Aguinaldo et al. 1997. Supcrphylum 1. Haemopoda superp. nov. (diagnosis: body segmented, with limbs on several segments; adult body cavity a hacmocoel that extends into the limbs). Phylum 1. Arthropoda von Siebold and Stannius 1848 (hard cuticle; jointed limbs moved by muscles). Subphvlum 1. Chcliceromorpha Boudraux 1978. Infraphylum 1. Pycnogonida Latreille 1810 (sea ‘spiders’, e.g.N ym phon). Infraphylum 2. Chelicerata Hcymons 1901 (arachnids, eurypterids, king crabs). .Subphylum 2. Trilobilom orpba Starmer 1944 stat. nov. (trilobites; trilobitoids). Subphylum 3. M andibulata Snodgrass 1938. Infraphylum 1. Crustacea* Pennant 1777 (e.g. barnacles, crabs, shrimps, , ostracods). Infraphylum 2. M yriapoda Leach 1814 (centipedes, millipedes, svmphylans, pauropods). Infraphylum 3. Insecta Linnaeus 1758 cm. auct. (Unnecessary junior synonym Hexapoda: e.g. springtails, s/lverfish, locusts, bees, , beetles, moths, Drosophila). Phylum 2. Lobopoda phyl, nov. (soft cuticle; unjointcd limbs with terminal claws; both muscles and hydraulic pressure involved in locomotion). Subphvlum 1. Grube 1853 (e.g. Peripatus). Subphylum 2. Tardigrada Doyèrc 1840 slat. nov. (water bears, e.g. Echiniscus). Superphyium 2. Nemathclminlhes superphyl. nov. (diagnosis: unscgmcntcd worms with spiny m outhparts; vascular system and limbs absent; coelom bounded by epithelium absent). Phylum Nemathclminthçs Gegenbaur 1859 em. (diagnosis as for the supcrphylum). Subphylum 1. Scalidorhvncha subphyl. nov. (diagnosis: retractile head covered with circlets ofspined sc a lid s). Infraphylum 1. Priapozoa infraphyl. nov. (diagnosis: unscgmcntcd; larva or adult with cuticular lorica of longitudinal plates into which it can retract: classes Priapula, Loricifera). Infraphylum 2. Reinhard 1887 (superficially segmented; without lorica). Subphylum 2. Rudolphi 1808 (as Nematoidea) orth. em. stal. nov. Infraphylum 1. Nematoda Gegenbaur 1859 onhog. cm. slat. nov. ie.g. Caenorhabditis, Ascaris). Infraphylum 2. Vcjovskv 1886 stal. nov. (e.g. Gordius, Nectonema). Infrakingdom 4. Platyzoainfraking. nov. [diagnosis: ciliated non-segmented acoelomates or pseudocoelomatcs lacking vascular system; gui (when present) straight, with or without anus]; possibly neotcnously derived from loxosomalid-like cntoproct larvae. Phylum 1. Acanthognatha phyl. nov. (diagnosis: chitinous cephalic bristles or jaws; gut if present with anus or anal pore). Subphylum 1. Trochata subphyl. nov. (diagnosis: syncytial epidermis with radial tubules penetrating the cuticle; often with paired lemnisci in neck region; epidermis multiciliatcd and/or noil-ciliated ; proboscis often functions as introvert; protonephridia flame bulbs; pscudocoelomatc, gonochoristic). Infraphylum 1. Rotifera* Cuvier 1798 stat. nov. (e.g. Collotheca. Asplancha). Infraphylum 2. Rudolphi 1809 slat. nov. (e.g. Moniliformis). Subphylum 2. M onokonta subphyl. nov. (without lemnisci or cvcrsible proboscis; monociliatcd epidermal cells; protonephridia solenocytcs; acoelomatc: herm aphrodite: classes Gastrotricha, Gnathostomulida). A revised six-kingdom system o f life 237

Table 4. (cont.)

Phylum 2. Platyhelminthes Gegenbaur 1859 cm. M inot 1876. Subphylum 1. Turbellaria! Ehrenberg 1831 em. Ehlers 1864 (usually multiciliated, non-syncviial epidermis retained in adult). Infraphylum I. M ucorhabda* infraphyl. nov. (mucoid non-lamcllatc rhabdoids, protonephridia absent or with two cilia, e.g. eatenulids, acoels). Infraphylum 2. Rhabditophora Ehlers 1985 (unranked) stat. nov. (lamellate rhabdilcs; protonephridia with many cilia: macrostomids, polyclads, Neoophora). Subphvlum 2. Neodermata Ehlers 1985 (unranked) stal. nov. (larval epidermis shed; replaced by syncytial neodermis). Infraphylum 1. Trem atoda Rudolphi 1808 slat. nov. Infraphylum 2. Ccrcomcromorpha Bychowsky 1937 slat. nov. (monogencans, tapeworms). Branch 2. Deuterostom ia Grobben 1908. Infrakingdom 1. CoeloraoporaM arcus 1958 (as supcrphylum) slat. nov. Phylum 1. Hemichordata Bateson 1885 stat. nov. orthog. cm. auct. Subphylum 1. Pterobranchia Lankester 1878 stat. nov. (e.g. Cephalodiscus', graptolites). Subphylum 2. Enteropneusta Gegenbaur 1870 slat. nov. (e.g. Balanoglossus). Phylum 2. ata De Brugicrc 1789. S u b p h y lu m 1. Homalozoa* W hitchousc 1941. Subphylum 2. Pelmalozoa Leuckart 1848. Infraphylum 1. Blastozoa Sprinkle 1973 (e.g. blasloids, cystoids). Infraphylum 2. Crinozoa M atsumoto 1929 (sea lilies, feather stars). Subphvlum 3. Eleutherozoa Bell 1891. Infraphylum 1. Asterozoa Von Zittel 1895 (starfish, brittle stars, eoncenlrieycloids). Infraphylum 2. Echinozoa Haeckel in Von Zitteli 1895 isea urchins; sea cucumbers). Infrakingdom 2. ChordoniaH a e c k e l 1874em. Hatschek 1888s la t. n o v . Phylum 1. Urochorda Lankester 1877 (urochordatcs). S u b p h y lu m 1. Tunicata Lamarck 1816itunicates). Infraphylum 1. Ascidiae Blainville 1824 (ascidians, sorberaccans). Infraphylum 2. Thaliae auct. (salps). Subphylum 2. Appendicularia Lahille 1890 slat. nov. (larvaccans). Phylum 2. Chordata Bateson 1885 em. Subphylum 1. Acraniala* Bleeker 1859. Infraphylum I. Cephalochordata Owen 1846 (amphioxus). Infraphylum 2. Conodonta Eichenberg 1930 stat, nov. (extinct). Subphylum 2. Vertebrata Cuvier 1812 (unnecessary synonyms: Craniota Haeckel, Craniata auct.). Infraphylum 1. Agnatha^ auct. (lampreys, ). Infraphylum 2. Gnathostomata auct. yawed fish, tctrapods). Subkingdom 4. M esozoasubking. nov. (bilateral multiccllular parasites with ciliated epithelium ; gut. nervous and vascular systems absent). Phylum Mesozoa van Beneden 1877 (dicyemids, orlhoncctids).

* Probably paraphyletic. t Almost certainly paraphyletic. The present system primarily emphasises classical morphological and cmbryological data and phylogenetic ideas, but has been revised lo take into account recent molecular phylogenetic evidence (reviewed by Ucshima, 1995; sec also Cavalier-Smith el a l., 1996a, Bridge et al., 1995; W innepenninckx et al., 1995; W innepenninckx, Backeljau and De W achter, 1995; Satoh, 1995; Mackey et al., 1996; Garev et al., 1 9 9 6 a, b)

Cercozoa is a new name for the assemblage of 1 9 95< j, b , 1997a). Because rRN A sequence has shown filóse and reticulose am oebae and phylogenetically that flagellate and amoeboid taxa are phylo­ related zooflagellates (including sarcomonads like genetically intermingled, the names Sarcodina and Cercomonas) for which the provisional name Rhizo­ Rhizopoda are now both abandoned as formal poda was recently adopted (Cavalier-Smith nam es for taxa. They will however rem ain useful as 238 T. Cavalier-Smith non-phyloger>et¡c designations of body form in plexan subphylum Apicomonada (Cavalier-Smith, descriptive and ecological studies, like ‘flagellate’ or 1993) are here transferred to the subphylum Pro­ ‘alga’ ; thus ‘rhizopod’ can continue to be applied in talveolata of the flagellate phylum Dinozoa, and the the traditional sense to any amoeba, irrespective of name Apicomplexa is abandoned as an unnecessary its taxonomic affinity, to contrast it with a flagellate junior synonym of the traditional name Sporozoa or sporozoan. Neomonada is a rather broad para­ [Corliss (1971) trenchantly and sensibly criticized phyletic assemblage with many phylogenetically the multidudinous unneccessary new names for the diverse lineages, some more closely related to one or S p o ro z o a sensu stricto]. Sporozoa in the present sense more of the four higher kingdoms of life than to each are restricted to obligate non-flagellate parasites, other. As our knowledge of neozoan phylogeny fin and unlike Apicomplexa sensu Cavalier-Smith which neomonad diversification played a central (1993c) include three taxa that lack an apical role) improves the classification and possibly also the complex : M anubrispora (comprising the metchniko- circumscription of the Neomonada will need to vellids, formerly regarded as microsporidians, but change: see Cavalier-Smith (1997 b). Though Cerco­ here excluded from Microsporidia because they zoa form a major clade on rRNA trees (Cavalier- appear to have a cortical alveolus like that of most Smith & Chao, 1997 and unpublished), some of sporozoans), Paramyxea, Haplosporidia. Further­ them are not phenotypically very distinct from some more, as outlined elsewhere (Cavalier-Smith, Neomonada, so the boundary between the two phyla 1998c), the parasitic flagellate P erkinsus is probably is not morphologically clear cut, and will probably less closely related to Sporozoa than are some free have to be reevaluated as our understanding of these living flagellates without an apical complex ; it centrally important but much neglected taxa im­ appears that its similarities to Sporozoa, which were proves. the reason for grouping them together in the new The class Athalamea was formerly grouped with phylum Apicomplexa (Levine, 1970) are conver­ Foraminifera in the phylum Granuloreticulosa (Lee, gent. No molecular evidence is yet available as to the 1990) or Reticulosa (Cavalier-Smith, 1993 c). Elec­ correct phylogenetic position of M anubrispora and tron microscopy ofP enardia (M ikrjukov & Mylnikov, Paramyxea, and that for the Haplosporidia is 1995) a n d Gymnophrys (Mikryukov & Mylnikov, ambiguous (Cavalier-Smith, 1997c). 1996 6) shows kinetocysts similar to those of some Cercomonas and a Golgi associated with the nuclear envelope as in several cercozoans and neomonads. Unlike Foraminifera they have no tests, their pseudopods do not show bidirectional streaming but (c) The infrakingdom Adinopoda differ from cercomonad pseudopods primarily in containing some microtubules, their trophic cells Though the monophyly of Heliozoa is rather have two subparallel centrioles with vestigial ciliary dubious, it would be premature to abandon either stumps, and they are freshwater, not marine. In the the taxon Adinopoda or the phylum Radiozoa. The absence of any specific ultrastructural similarities suggestion that Radiozoa are polyphyletic (Amaral with Foraminifera, the taxon Granuloreticulosa or Zettler, Sogin & Caron, 1997) may be mistaken: the Reticulosa appears unnatural. Because of their failure of Acantharia and Radiolaria to group similarities to cercozoan amoeboflagellates, Atha­ together on their distance trees could simply be an lam ea (sole order Biomyxida) are here transferred to artefact of the especially high evolutionary rate of the subphylum Monadofilosa of the Cercozoa, even rRNA in Radiolaria, as clearly indicated by their though they have flat mitochondrial cristae whereas long branches, which probably places them too low most cercozoans possess tubular cristae. Foramini­ in the tree, an artifact that is even more marked for fera thus become a phylum in their own right, Mycetozoa and Microsporidia. M y own unpublished characterized by granuloreticular pseudopods ema­ maximum likelihood analyses indicate that Radio­ nating from one or more pores in their tests and laria and Acantharia may actually be sister groups tubular mitochondrial cristae; Granuloreticulosa/ and that Radiozoa is probably a valid phylum; the Reticulosa are discontinued. position of Radiolaria, however, is not robust and with some taxon samples they have a tendency to associate instead with other long branches, but not lb) The infrakingdom Alveolata always the same ones as observed by Amaral Zettler, The flagellates formerly grouped as the apicom- Sogin & Caron (1997). A revised six-kingdom system o f life 239

V. THE KINGDOM ANIMALIA AND ITS 23 flagellate and the first , during which colla­ PHYLA genous mesenchymal connective tissue and epithelia first evolved, it is highly improbable that Mesozoa The animal kingdom is here divided into only 23 are phyletlc links between Protozoa and Animalia. phyla, mostly familiar but a few novel, grouped in As discussed in section IV molecular evidence four morphologically very distinct subkingdoms indicates that mesozoans are derived from Bilateria (T a b le 4). by the loss of nervous system and gut. Because of these radical differences I continue to treat them as a separate subkingdom, even though this makes (1) Radiata, the ancestral animal subkingdom Bilateria paraphyletic. su b k in g d o m

I have used the more ancient term Radiata of (3) The number of animal phyla Lamarck and Cuvier rather than the more recent Diploblastica (Haeckel, 1866) for the subkingdom Hyman (1940) recognised only 21 animal phyla or comprising the four most primitive animal phyla, later 23 (1959). Many American college textbooks even though R adiata originally included also Echino­ now recognize about 35, and Möhn (1984) had as dermata and often other taxa which are now many as 38. This taxonomic inflation has mainly properly classified within Bilateria. The term Diplo­ come about not by the discovery of new groups or by blastica is entirely inappropriate for this subkingdom a clear demonstration that many of Hym an’s phyla since the majority of Radiata are not diploblastic, were polyphyletic and had to be split. Much of it but triploblastic. Only the cnidarian class Hydrozoa arose by the excessive splitting of the phyla Arthro­ is truly diploblastic. The fact that the cnidarian poda and Aschelminthes of Hyman (1940); in the subphylum Anthozoa (i.e. the majority of Cnidaria) case of the every class has been treated is fundamentally triploblastic was emphasized by as a separate phylum, not because of convincing Pantin (I960), while the mesogloea of Scyphozoa evidence that they are phyletically unrelated but also frequently contains cells as does that of merely because their phyletic relationships have Ctenophora (Hyman, 1940). Sponges are funda­ been unclear. In essence phylogenetic agnosticism, mentally triploblastic, as their mesohyl contains rather than reasoned arguments, has led to an many cells. As sponges are almost certainly the first unnecessary inflation in the num ber of phyla, which animals and probably arose directly from choano- obscures rather than clarifies their evolutionary flagellate protozoa (Kent, 1881; Cavalier-Smith, relationship and makes it more difficult for students 1981 a; W ainright el a l., 1993; Cavalier-Smith et at., to appreciate the diversity of animals than would a 1996a), the kingdom Animalia is ancestrally triplo­ system with fewer phyla. In the present system I blastic (Cavalier-Smith, 1983a) and the diploblastic have therefore taken the bull by the horns — or condition of Hydrozoa alone is a derived simplifi­ perhaps I should say the priapulid by the spines cation. Haeckel’s (1866) influential idea that triplo- and grouped the classes into just two blasty evolved from diploblasty is clearly wrong and, phyla, each of which I think is monophyletic. As a to use Hyman’s (1959) trenchant phrase (when result of also adopting a broader definition of the referring to another phylogenetically incorrect con­ Annelida and merging brachiopods and phoronids cept of Gephyrea), ‘must be obliterated from into a single new phylum, Brachiozoa, my present zoology’. Though the term diploblastic may prop­ system still has only 23 phyla, even though (unlike erly be applied to Hydrozoa, radiates should no Hyman) I accept the phylum Placozoa, treat longer be referred to as ‘diploblasts’ (e.g. Christen et Urochorda as a separate phylum from Chordata, a i , 1991; Ueshima, 1995). and segregate and onychophorans from As the mesohyl of sponges is probably both Arthropoda as the phylum Lobopoda. I d o n o t homologous with and ancestral to the mesogloea of accept the recently discovered Cycliophora (Funch Cnidaria it would be sensible to call it mesogloea & Kristensen, 1995) as a separate monogeneric also. p h y lu m ; Sym bion is sufficiently similar in body plan to Entoprocta to be grouped with them in an emended phylum Kamptozoa, but in a separate (2) The derived subkingdom Mesozoa subphylum. Thirteen of the present animal phyla Since the kingdom Animalia almost certainly are identical to those of Hyman, and seven are evolved by a transition between a colonial choano- identical with those of the 18 phylum system of 2 4 0 T. Cavalier-Smith

Lankester (1911 i, though I use the older name dealing with the echiuroids and and Annelida in preference to his Chaetopoda. therefore gave no more detailed discussion than that, her unsound decision has been widely followed by textbooks. However, the fact that echiuroids have (4) A broadened phylum Annelida well developed chaetae, typical trochosphere larvae, The phylum Annelida recognised here is that of and a vascular system and nephridia similar to those Sedgewick (1898) with the addition of the Pogono­ of other annelids, leaves little doubt that they are phora, which were then unknown. Because of their annelids in which the formerly discrete coelomic anterior oligomerous coelom and the belief that they cavities have become secondarily merged into one. lack chaetae, early workers on Pogonophora thought As Sedgewick pointed out, there are traces of they were unrelated to annelids and c lo s e r to segmentation in young Echiurus. Loss of septa and , and so deserved their own phylum partial merger of coelomic cavities is very common (Reisinger, 1938). Hyman (1959) accepted this, but in polychaetes and virtually complete merger oc­ at that time it was still not known that early workers curred in leeches (ClaTk 1964); the almost total were completely ignorant of the posterior segmented suppression of segmentation in adult echiuroids is chaetiferous portion of the anim al’s body. Since this probably a locomotory adaptation to a relatively was discovered it has been clear that Pogonophora sedentary burrowing life in which the proboscis took are not deuterostomes, but instead are related to on an even more important role than in analogous annelids. The closeness of this relationship has been burrowing polychaetes. These adaptations do not strongly confirmed by sequence data (Winnepen­ constitute a fundamentally different body plan from ninckx, Backeljau & De W achter, 1996). It is now other annelids. Treatment of the small number of evident that Pogonophora are annelids that have Echiuroida as a subphylum of the Annelida is lost their gut as a result of the evolution of the greatly preferable to making them a phylum in their capacity to cultivate chemosynthetic bacteria within own right, which obscures their true affinities, and their body and give up predatory feeding. They devalues the idea that separate animal phyla should probably arose from tubicolous polychaetes. Though have fundamentally different body plans. loss of the gut and evolution of the ability to cultivate bacteria in a highly developed trophosome are (5) The pseudocoelomate phyla important innovations, Pogonophora retain such a Nemathelminthes and Acanthognatha phyl. major part or the fundamental body plan n o v . (segmentation with chitinous chaetae, haemoglobin- containing vascular system and respiratory tentacles The proper status of the pseudocoelomate aschel- like polychaetes) that there is no longer any minths has always been problematic. But as, Loren- justification for treating them as a separate phylum, z e n (1985)stressed, there are many synapomorphies still less in dividing them into two phyla as has been that can be used to establish real relationships done. Ranking Pogonophora as a subphylum (within between some of the classes. Therefore it is highly which typical pogomophorans and Vestimentifera unsatisfactory to rank almost every class as a separate can be separate classes) is sufficient recognition of phylum. But, like most zoologists, I do not think that their distinctiveness from Polychaeta and Clitellata, aschelminths as a whole are monophyletic. Instead both here ranked as subphyla of the Annelida. there seem to be two fundamentally different Sedgewick (1898) regarded the class Echiuroidea phylogenetic lineages, which I treat as phyla, and a as ‘obviously true Annelids’. He was the first to larger num ber of goupings of related classes, which I realize they were radically different from sipuncu- treat as subphyla and infraphyla. Some of these are loids, with which they were previously grouped in more strongly supported by existing data than the unnatural phylum Gephyrea (which Hyman o th ers. rightly ‘obliterated from zoology'); he first made The new subphylum Trochata (comprising roti­ Sipunculoidea a separate phylum. Hyman (1940) fers and Acanthocephala) is very strongly supported first treated Ecbiuroida as a separate phylum on the by the synapomorphies listed in Table 2 [sec also grounds that they are ‘unsegmented', whilst at the Lorenzen (1985) and Nielsen (1995)] as well as by same time admitted that they might alternatively be molecular trees ( W innepenninkx e t a i , 1995 a; Garey appended to the Annelida, ‘ to which phylum they et al. 19966). Ranking Acanthocephala as a phylum are undoubtedly related’ (Hyman, 1959). Even (Hyman, 1940) overemphasized the importance of though Hyman never wrote a volume of her treatise the secondary loss of the gut and cilia; these A revised six-kingdom system o f life 241 degeneracies, the syncytia! epidermis, and proboscis their own phylum. Therefore I rank both groups as hooks are adaptations to parasitism remarkably subphyla within the new phylum Brachiozoa. convergent with those of cestodes, which are also not Though molecular data support the relationship to placed in a separate phylum from their closest free- of both Onychophora and Tardigrada living ciliated relatives, the Turbellaria-1 rank both (G a re y et a l., 1996a), I separate them from Arthro­ Rotifera and Acanthocephala as infraphyla so that poda as a new phylum Lobopoda (a name proposed Bdelloidea and Monogogonta can be classes, as informally by M antoni 1977) because unlike true preferred by those who have treated Rotifera as a arthropods they do not have jointed limbs and a phylum. Even though Acanthocephala probably rigid cuticle. arose from rotifers by parasitic degeneration, I prefer not to place them within Rotifera as suggested by G a re y et al. (1996é), since the two groups are (7) Phylogenetic assum ptions behind the phenotypically so different and it would be nomen- new bilaterian infrakingdoms and claturally confusing to call Acanthocephala rotifers su p e r p h y la merely because they evolved from them. The new subphylum Monokonta (comprising Even if we adopt a broad phylum Annelida and gastrotrichs and gnathostomulids) is primarily based recognize only two aschelminth phyla, there are still on their monociliated epithelial cells (whence the 17 separate bilaterian phyla, which fall naturally name) and needs testing by gene sequencing. The into two branches: Protostomia (13 phyla) and grouping of Monokonta and Trochata as the new Deuterostomia (4 phyla). Deuterostomia are divis­ phylum Acanthognatha also needs such testing, as it ible into two long-standing groups, Coelomopora is not clear whether the chitinous jaws of rotifers and (Hemichordata and Echinodermata) and Chordo- gnathostomulids are homologous or not. nia (Urochorda and Chordata), which I h e re ra n k The grouping of Nematoda and Nematomorpha as infrakingdoms, though a rank of superphylum as subphylum Nematoida is an old idea, well would also be acceptable. For , however, supported both by morphology (Lorenzen, 1985) the number of phyla is so large that we need both and gene sequences (Winnepenninkx et a l., 1995a). infrakingdoms and superphyla in order to group The similarities between the lorica of longitudinal them informatively. I divide them into four new plates in Loricifera and larval priapulids is the basis infrakingdoms. Ecdysozoa comprise the three phyla for grouping them as the new infraphylum Priapo- (Arthropoda, Lobopoda, Nemathelminthes) with a zoa. Priapozoa arc grouped with kinorhynchs as the moultable cuticle and a very poorly developed or new subphylum Scalidorhyncha, since they all have absent coelom (Aguinaldo et a l., 1997). As they are a retractile head covered with circlets of spined undoubtedly more closely related to each other than scalids, which seems unlikely to be convergent. to the Nemathelminthes 1 group Lobopoda and M olecular testing is however needed- My grouping Arthropoda together as the new superphylum of Scalidorhyncha and Nematoida as the Nema- Haemopoda, because of their shared segmented thelminthes (an old phylum name) might be body and haemocoel that extends into the limbs. questioned on the ground that such a grouping was This superphylum enables one to indicate that not evident on a recent rRNA tree (Winnepenninkx lobopods are closely related to, but not actually et al., 1995a). However that tree does not specifically arthropods. By adopting the name Lobopoda I do relate either Nematoida or priapulids to any other not imply any connection with polychaetes, some of group, and thus lacks resolution as to their re­ which have been referred to as lobopodial (Sharov, lationship and therefore does not clearly disprove the 1966). 1 do not think there is anv direct phylogenetic monophyly of the Nemathelminthes postulated here. connection whatever between Haemopoda and Annelida; haemopod limbs and polychaete para- podia have always seemed to me merely analogous, (6) The new phyla Brachiozoa and Lobopoda not homologous as assumed by the annelid theory of Recent rRNA phylogenies show that phoronids the origin of arthropods (Snodgrass, 1938; Sharov, probably arose from inarticulate brachiopods by the 1966). The gulf between Ecdysozoa and Lophozoa, loss of the shell (Cohen, Gawthrop & Cavalier- the other major infrakingdom, is im­ Smith, 1997). Since in other respects both groups mense and very difficult to cross in plausible share a basically similar body plan, phoronids are to megaevolutionary steps, which is why I have given brachiopods as slugs arc to snails, and do not meric both the high rank of infrakingdom. 242 T. Cavalier-Smith

Lophozoa, the largest protostome infra­ and that each of these groups became cephalized kingdom, comprise the six protostome phyla that independently. Given that both radiate phyla have well developed coelomic cavities plus the (Cnidaria and Ctenophora) are uncephalized, there Kamptozoa, which do not. Unlike most zoologists, 1 is no phylogenetic reason to think that the ancestral do not accept Hym an's (1951) view that the absence bilaterian was cephalized. I argue that all Lophozoa of a coelom in entoprocts precludes a specific are ancestrally non-cephalized and that the Bivalvia relationship with Bryozoa. Like Nielsen (1995) I among the molluscs and the tubicolous Operculata think the two groups are related. I suspect that among the annelids are closer to the ancestral state Kamptozoa arose from an early bryozoan by the loss than their more cephalized relatives. The bivalved of the coelom. Though I do not altogether rule out character of the polyzoan larvae, of brachiopods, the reverse possibility that the coelom first evolved in and bivalve molluscs suggests that the common Bryozoa from a non-coelomate entoproct-like an­ ancestor of Polyzoa and Conchozoa had a bivalved cestor, I prefer the view that Bilateria were ances­ larva and sedentary non-cephalized adults. On this trally coelomate and that the coelom first arose view the traditional creeping radulate archimollusc simultaneously with the bilaterian gut by partition­ (von Salvini-Plawen, 1990) was not the ancestor of ing the anthozoan coelenteron into two parts, as Mollusca as a whole, but just of the subphylum proposed in my invited manuscript for the System- Glossophora (i.e. the radulate molluscs). atics Association 1984 meeting on the relationships The other novel assemblage, the superphylum of lower invertebrates that was rejected by the Vermizoa (a name I proposed at the 1984 meeting) editors because it was too controversial (Conway comprises Annelida and Nemertina, both of which M o rris et a l., 1985). My novel version of the have well developed closed vascular systems. Though Archicoelomate theory (Ulrich, 1949) of the origin it is sometimes denied that the rhynchocoel of the and diversification of bilateral animals was clearly nemertines is a true coelom, the presence of the unacceptable to zoologists raised on Hyman’s dog­ vascular system and gut with an anus makes them matic view that all non-coelomates were ancestrally much more similar to annelids than to the flatworms, so. My present grouping of both Bryozoa and with which many authors have grouped them. Since Kamptozoa in the superphylum Polyzoa may be echiuroids testify to the likelihood that some annelids equally unacceptable to many. But I have yet to see can lose segmentation, whereas in others the coelom a more convincing explanation of the origin of can become occluded by parenchyma (Rieger, Kamptozoa than by coelomic reduction, possibly 1985), there are no strong morphological reasons from a lophopod-like bryozoan. against grouping annelids and nemertines in the Like the resurrection of the Polyzoa, my grouping same superphylum. of the Brachiozoa with the Mollusca as the super- Ribosomal RNA trees show that all the lophozoan phylum Conchozoa may seem a retrograde step, phyla, except the Sipuncula, are so closely related to harking back to Cuvier’s inclusion of brachiopods in each other that their branching order cannot be the Mollusca. How'ever, as I argued at the 1984 readily determined. Because of their unique charac­ meeting, Hyman's dogma that Polyzoa, Brachiozoa ter Sipuncula are placed solitarily in their own and Bivalvia are decephalized is probably incorrect. lophozoan phylum. The name Lophozoa was chosen There is no phylogenetic evidence whatever that any on the assumption that the ancestor had a retractile of these animals ever had heads. Hyman’s dogma pair of tentacles with an interior coelomic cavity. was expressed so forcefully as to inhibit dissenting This is generally agreed for the Bryozoa and thought: ‘lophophorates and deuterostomes as seen Brachiozoa, but the ctenidia of molluscs, the tenta­ today are all decephalized animals of sedentary or cles of Sipuncula, and the gills of operculate sessile habits. It is inconceivable that such types polychaetes may, I suggest, also be homologues of should originate the Bilateria. It appears self-evident the . It is much more doubtful that the that only well-cephalized, active forms can originate nemertine proboscis has a similar origin, but the definitive bilateral symmetry’. Her view was purely possibility cannot be totally dismissed. If the enio- intuitive, unsupported by any phylogenetic reason­ procts are derived from Bryozoa by coelomic ing. The fact that the heads of radulate molluscs, occlusion their tentacles also may be derived from annelids, nemathelminths, arthropods and verte­ those of the lophophore. brates are all so different from each other, and lack The deep divergence between the Ecdysozoa and any recognizable homology, argues strongly that the Lophozoa is supported by the rRNA trees, which their common ancestor was not strongly cephalized are also inconsistent with an annelid ancestry for A revised six-kingdom system o f life 243 arthropods and Cuvier’s old group Articulata, which easily swamp the true phylogenetic signal and yield is clearly poylphyletic. Since arthropods and mol­ a positively incorrect answer. luscs appear almost simultaneously at the beginning I do not of course suggest that Ecdysozoa or of the early Cambrian bilaterian fossil record, evolved directly from any extant polyzoan Ecdysozoa and Lophozoa probably diverged very- group. The discovery of Sym bion suggests that the soon after the origin of the Bilateria. Any theory of early radiation of Kamptozoa may have been much how they are related is necessarily highly speculative. more varied than we can now guess from the few The speculation that I presented at the 1984 extant species. Unfortunately the fossil record of this Systematics meeting, and which I still favour was phylum of tiny animals is so poor that it will that haemopod limbs evolved from tentacles of an contribute very little to testing my thesis that similar early solitary polyzoan. I had in mind the solitary organisms could have been ancestral to both Ecdyso­ loxosomatid entoprocts, which can locomote actively zoa and Platyzoa. on the tips of their tentacles (Hyman, 1951). The The fourth protostome infrakingdom is the pscudocoel of such an active entoproct could have Chaetognathi, which are so different from all other given rise directly to the haemocoel of the Haemo­ protostome animals that it does not seem possible to poda and the pseudocoel of Nemathelminthes. Such group them with any other phylum. Their rRNA an origin involves much less change than the view sequences also leave them in an isolated position, that arthropods evolved from a coelomate legless and give no support to Hyman’s view that they are worm. Entoproct tentacles can already function deuterostomes. The protostome infrakingdoms similarly to lobopodal limbs and are similarly might be further reduced to three by placing arranged in two lateral rows. Nemathelminthes chaetognaths in the Ecdysozoa, with which they could have arisen from an early lobopod through the share some characters, but I prefer to emphasize loss of limbs after adopting a burrowing habit, like their uniqueness. snakes, apodans, and caecilians among vertebrates. The third protostome infrakingdom is the non- (8) New animal subphyla and infraphyla coelomate Platyzoa, which contains only the phyla Acanthognatha and Platyhelminthes. Both are an­ Subphyla have long been used in the Arthropoda cestrally slender ciliated worms with no vascular and Chordata which have the largest number of system, either acoelomate or pseudocoelomate. disparate classes. In other phyla I have made more Though Hyman (1940-1959) was utterly confident extensive use of them than is traditional, partly to that they represent the ancestral bilaterian con­ avoid unnecessary splitting into separate phyla and dition, I am deeply sceptical of this, as I cannot partly to emphasize that some classes really are more envisage how they might have evolved from a closely related to each other than to others. Most of radiate ancestor. It seems to me more likely that they the subphylum groupings are well accepted and I arose from neotenous entoproct larvae of the have often used traditional names; but slight changes creeping type found inLoxosom a: even Hym an (1951) in rank are frequent (dowmward from phylum and noted their remarkable similarity to a . If such upward from class or superclass). Only three an origin is correct, then no separate coelomic loss subphyla are new both in concept and name: need be postulated. An alternative way of sub­ Scalidorhyncha (priapulids, loriciferans and kino­ dividing the Protostomia to that in Table 4 would be rhyncha); Trochata (rotifers and acanthocepha- to transfer Kamptozoa from the Lophozoa to the lans) ; Monokonta (gastrotrichs and gnathosto­ Platyzoa. Though compatible with my phylogenetic m u lid s). assumptions, I prefer to emphasize the life cycle and In phyla with a large number of classes I h av e trophic similarities of Polyzoa, rather than the loss of used infraphyla much more extensively than usual. the coelom. The rRNA trees are unfortunatly not Again, some of these, like Cephalopoda, Agnatha very helpful in understanding the affinities of the and Rotifera, are well known taxa given inter­ Platyhelminthes; their branches (like those of nema­ mediate ranks in order to include more cladistic todes) are so long that one must suspect that long- information in the classification. Only six are new in branch artefacts may be placing them lower in the both concept and name: the radulate mollusc tree than their correct position. When one has a infraphyla Monoconcha (those with a single shell) rapid, almost simultaneous radiation, as appears to and Spiculata (those with multiple shell plates be the case, coupled with an elevated evolutionary and/or calcareous spicules) ; the polychaete Opercu­ rate in some branches, such systematic biases may lata and Pharyngata; Priapozoa (priapulids and 2 4 4 T. Cavalier-Smith loriciferans) ; and M ucorhabda (turbellarians with mirnics', protists that have convergently become non-lameilate rhabdoids). Elementary texts would similar to fungi in one or more respects. benefit from using subphyla more extensively than The protozoan groups (slime moulds and plasmo- they do, but infraphyla may not be necessary for diophorids) remain vegetativeiy amoeboid but have them, though will be valuable in more specialized acquired aerial spores for dispersion convergently works and databases. with fungi. Slime moulds are polyphyletic, but the majority of them form a monophyletic protozoan infraphylum Mycetozoa (Table 2) (i.e. classes VI. THE KINGDOM FUNGI AND ITS FOUR Protostelea, M yxogastrea and Dictyostelea) ; though PHYLA the myxogastrean Physarum polycephalum a n d th e Dictyostelium discoideum most often do not (1) Circumscription of the Fungi group together on 18S rRNA distance trees they A kingdom Fungi entirely distinct from the kingdom usually do so on 18S rRNA maximum-likelihood Plantae is now almost universally recognized {Whit­ trees (e.g. Cavalier-Smith, 1995a, 6), which appear taker, 1969; Carlile & Watkinson, 1994). Recently to be less misled by the very long Physarum b ra n c h . this kingdom has sometimes been referred to instead Inspection of the alignment shows about seven as Mycota, but this name has probably not been obvious molecular synapomorphies for the Myceto­ validly published and is entirely unnecessary and zoa, and Rusk, Spiegel & Lee (1994) have rRNA much less widely understood. It is increasingly clear sequence evidence that the protostelids are specifi­ that Fungi are more closely related to animals than cally related to and myxogastreans, to plants (Baldauf & Palmer, 1993 ; W ainright et al., which is also supported by trees based on sequences 1993); however, it is incorrect to speak of an of the protein elongation factor EF la (Baldauf, 'evolutionary link' between fungi and animals & Doolitde, 1997). Molecular evidence is not yet (W ainright et a l., 1993) since the phylogenetic available for the minor groups of non-mycetozoan linkage is indirect via the choanofiagellate protozoa. slime moulds, the acrasids, copromyxids and fonti- As Cavalier-Smith (19876) first suggested, it seems culids. However, the morphology of their mito­ that Fungi evolved from choanofiagellate protozoan chondria and pseudopodia convincingly places ancestors as did animals, but independently. Acrasida within the class Heterolobosea (Page, 1987) M olecular evidence strongly supports the restric­ within the protozoan phylum Percolozoa (Cavalier- tion of the kingdom Fungi to the taxa summarized in Smith, 1993a, c). While their morphology is less Table 5, including (as discussed in section III above) decisive, copromyxids are placed in the protozoan the unexpected inclusion of the Microsporidia. The class Lobosea of the phylum Amoebozoa, whilst earlier circumscription that I advocated previously fonticulids are in the class Cristidiscoidea, now excluding the Microsporidia (Cavalier-Smith, within the subphylum Choanozoa of the protozoan 1981a, 19876) was supported by much rRNA phylum Neomonada (Cavalier-Smith, 1998a; for­ sequence evidence (Bruns, W hite & Taylor, 1991), merly Cristidiscoidea were in the Rhizopoda : Page, and accepted by the authoritative Dictionary o f the 1987; Cavalier-Smith, 1993a, 19956, 1997a); and F ungi (Hawksworth et a i , 1995). Several other taxa there is no reason whatever to group them with have often been treated as Fungi, but are evoludon- Fungi. Plasmodiophorids have often been thought to arily quite unrelated to them. Some recent be related to Mycetozoa and therefore were treated texts (e.g. Alexopoulos, Mimms & Blackwell, 1996) as fungi; but molecular evidence clearly shows that correctly exclude these elements from the kingdom they are protozoa that belong in the Cercozoa Fungi, whereas others (e.g. Carlile & Watkinson, (formerly Rhizopoda: Cavalier-Smith & Chao, 1994) still incorrectly place them in the same 1997). k in g d o m . In contrast to the foregoing protozoan taxa, The most important of the non-fungal groups oomycetes, hyphochytrids and thraustochytrids all widely studied by mycologists and therefore formerly belong in the kingdom Chromista (see section V III classified as Fungi are the slime moulds, plasmo- below). They are probably all secondarily non- diophorids, oomycetes, hyphochytrids and thrausto- photosynthetic heterotrophs that have acquired chytrids, which all properly belong either in the vegetative cell walls convergently with fungi. kingdom Protozoa or in the third botanical kingdom, The protozoan Corallochytnum limacisporum h as Chromista, as is fully accepted by Hawksworth et al. been erroneously treated as a (Raghu- (1995). These organisms are not fungi but ’fungus- Kumar, 1987); however, rRNA phylogeny reveals A revised six-kingdom system o f life 245 that it acquired a cell wall and evolved osmotrophy long caused some zoologists to be sceptical of the idea entirely independently of Fungi and heterotrophic that they are primitive eukaryotes. An origin from chromists, and is actually a protozoan related to one of the trichomycetes, which are all obligate choanoflagellates (Cavalier-Smith & Allsopp, 1996). endoparasites of arthropods does not suffer from this Other Protozoa, notably parasitic Ichthyosporea problem and makes sense ecologically as well as (now in subphylum Choanozoa: Cavalier-Smith, phylogeneticallv and cytologicallv (Cavalier-Smith, 1998a) have also erroneously been treated as fungi in 1998*). the past, whereas some Fungi, notably Pneumocystis (3) Revision of higher fungal classification (E d m a n el a l., 1988), have mistakenly been thought to be Protozoa. The classification in Table 5 is based on that of Cavalier-Smith (1987Ä) but modified in several im portant respects, including the subdivision of the (2) The trichomycete origin of microsporidia kingdom into the two subkingdoms Eomycota and The key step in the origin of microsporidia apart Neomycota. The new names of higher fungal taxa in from the losses of these two organelles was the origin Cavalier-Smith ( 1987 é) were at the editor’s in­ of intracellular parasitism and of the polar tube, the sistence not accompanied by proper diagnoses. This organelle that uniquely distinguishes the phylum is rectified in Table 5 which validates them and all from all others. The polar tube is formed within the new fungal names introduced here for the first time. spore from membranes thought to be of Golgi Since many mycologists have in recent decades been character, and is typically long and highly coiled. In curiously reluctant to use higher taxa than orders I germinating spores it is extruded and its apex have thought it desirable to make the present fungal attaches to host cells enabling the sporoplasm to classification more comprehensive than for the other enter the host cell through the tube. Clearly the five kingdoms. I therefore include superclasses, origin of the polar tube and this unique method of classes, subclasses and superroders in order to bring infection was both necessary and sufficient for the fungal megasystematics more into line with the origin of intracellular parasitism and the first practice in other kingdoms. microsporidian. A necessary corollary of this type of The most important innovations are the creation injection mechanism is the of the spore wall of the new hemiascomycete class Geomycetes and a and the temporary nakedness of the infective stage. major revision of the phylum Archemycota, which Temporarily naked infective cells are common in includes not only the fungi placed by Bruns, W hite Chytridiomycetes, the most primitive fungal group. & Taylor (1991) and many other authors in the Furthermore chitin walls have secondarily lost, nomenclaturally invalid phyla and apparently independently in the vegetative cells of , but also the Laboulbeniales, which the allomycete Coelomomyces and the trichomycete were traditionally regarded as ascomycetes.I a rg u e d Amoeboidium, which both parasitize animals. Thus previously (Cavalier-Smith, 1987ó) that ‘Zygo­ the secondary loss of chitin walls in fungal parasites m ycota’ do not differ sufficiently from ‘Chytridio­ of animals is not unprecedented and could also have mycota’ to merit a separate phylum. I am unaware occurred in the postulated fungal ancestor of of any mycologist who has seriously argued the case microsporidia and thus preadapted it to the origin of for having two separate phyla: nonetheless this the polar tube and an intracellular mode of life. The taxonomic inflation has been widely adopted in the intracellular stage of microsporidia is not always past decade. entirely naked, but sometimes bears a surface coat, I suggest that traditional trichomycetes (Licht- w h ic h in Pleistophora is so thick as almost to resemble wardt, 1986), which are all gut parasites of arthro­ a w all. pods, are polyphyletic. Two orders (Eccrinales and Elsewhere I propose that the microsporidian polar Amoebidiales) have Golgi dictyosomes, unlike all tube may have evolved from apical spore bodies, other Zygomycotina, so I remove them from the which are organelles of certain harpellalean Trichomycetes and Zygomycotina as a new class trichomycetes, probably extrusive, that are thought Enteromycetes. I had earlier (Cavalier-Smith, to help the spore attach to the gut lining of their 1981a) divided the traditional Chytridiomycotina host (Cavalier-Smith, 1998Í). The fact into two classes: Chytridiomycetessensu stricto (w ith that microsporidia mainly infect invertebrates, es­ Golgi dictyosomes) and Allomycetes (without dictyo­ pecially insects, and only very rarely infect protozoa, somes), both of which I validate in Table 5. I group and are unknown as parasites of Metamonada has Chytridiomycetes and Enteromycetes together to 246 T. Cavalier-Smith

T a b le 5. Classification of the kingdom Fungi, its four phyla and 20 classes

Subkingdom1. E o m y co ta |subking. nov. (hyphac, when present, usually lacking septa or rarely with imperforate septa; absent; vegetative walls frequently absent in animal parasites; mitochondria and peroxisomes often absent, sometimes replaced by hydrogcnosomcs: ccllulac non binueleatae; septa usitale absens; si praesens non perforata). Phylum 1. Archemycotafphyl. nov. (diagnosis: mitochondria and peroxisomes usually present; if absent then possessing hydrogcnosomes; vegetative cell walls usually present: mitochondria aut hydrogcnosomac praesentes. Name introduced without Latin diagnosis by Cavalier-Smith, 19876). Subphylum 1. Dictyomycotinaf subphyl. nov. (diagnosis: Golgi dictvosome of slacked cisternae: dictyosoma p r a c b c n s ). Class 1. Chytridiomycetes De Bary 1863 stat. nov. Sparrow 1958 em. Cavalier-Smith 1981 (emended diagnosis: posteriorly ciliated zoospores: sporae cilus posterioribus inslructae). Subclass 1. Rumpomycctidae subcl. nov. (diagnosis: zoospores with rumposome: zoospora rumposoma instructa; introduced as a class name without Latin diagnosis by Cavalier-Smith, 19876) (orders Chytridiales, M onoblcpharidalcs). Subclass 2. Spizomycctidac subcl. nov. (diagnosis: rumposome absent: sine rumposoma; introduced as a class name without Latin diagnosis by Cavalier-Smith, 19876) (orders Spizellomycetales, Ncocallimasiigalcs). Class 2. Enteromycetes cl. nov. (diagnosis: cilia absent; parasites of arthropod gut: sine ciliis; intestinum arthropodis incolens) (orders Eccrinales, Amocbidialcs). Subphylum 2. M elanomycotina subphyl. nov. (diagnosis: Golgi cisternae unstackcd; melanin-pigmented resting spores common; mitochondria and peroxisomes present: dictyosoma non-pracbcns: cisternae disjunctae; sporae saepe luscae). Infraphylum 1. Allomycotina infraphv. nov. (diagnosis: with uniciliate zoospores.: zoospora cilio único instructa). Class 1- Allomycctcs cl. nov. (diagnosis: with uniciliate zoospores: zoospora cilio único instructa; name introduced without Latin diagnosis by Cavalier-Smith, 1981 a) [orders Blastocladialcs, Coelomomycetales ord. nov. (diagnosis: trophic phase without walls: sine muris in statu pabulalorio; sole Coelomomycetaccac)]- Infraphylum 2. Zygomycotina! infraphyl. nov. (diagnosis: cilia and zoospores absent: sine ciliis). Superclass 1. Eozygomycctia supcrcl. nov. (diagnosis: without sporangiosporcs or aquatic spores, saprophytes or symbionts of vascular plants: sine sporangiosporis aut sporis aquaticis). Class 1. Bolomycetes cl. nov. (diagnosis: saprophytes with 11—12-singlel centriole; single large propulsive conidia borne on long unbranchcd conidiophores; zygospores with adjacent beak-like projections; not within a sporocarp; septate mycclia: conidophora instructa; centriolum microtubulos continens) [Basidiobolales ord. nov. (diagnosis as for Bolomycetes) [Basidiobolus]]. Class 2. Glomomycetes* cl. nov. (diagnosis: form vesicular arbuscular endomycorhizas with vascular plants; sclcrotium-like sporocarps contain chlamydosporus (Glomalcs) or laterally produced zygospores (Endogonalcs) ; centrioles, conidia, aerial spores and stalked sporophores absent. Conidiophora el centriolum absens; in radices plantarum crescens; sporae in sporocarpiis subterraniis). Superclass 2. Ncozygomycctia supcrcl. nov. (diagnosis: sporocarp and centrioles absent; aerial or aquatic aplanospores; saprophytes or parasites ot animals: sporangiospora aut sporae aquaticae instructa). Class 1. Zygomycetes cl. nov. (diagnosis: zygospore wall modified from the gamelangial wall, produced between the gam clangia; passive aerial asexual spores borne on stalked sporophores; usually saprophytes: mttrtis gametangii in murum zygosporae transiens; sporae aeriae in sporophoro pedicellata). Subclass 1. M ucoromycetidae Fries 1832 slat. nov. (diagnosis: coenocvtic; septa absent in young m yrella; with globose multispored sporangia and/or uni- or oligo- sporic sporangiola; sporophorc hypha-likc) [orders M ucorales: sporangium with columella; zygospores lacking an investment of sterile hyphac; Mortierellales ord. nov. (diagnosis: sporangium without columella; zygospores often invested by sterile hyphae; chlamydosporus: sporangium sine columello; hyphae steriles zygospora saepe inveslientes; chlamydospora inslruclalcs)]. Subclass 2. M cromycctidae subcl. nov. [diagnosis: aerial spores formed in merosporangia or as conidia; globular sporangia absent; mycelia septate with medial plug (Dimargarilales, Kickxcllalcs) or not (Piptocephalaceae. Cuninghamcllalcs) ; sporophorc often vesicular: sporae aeriae in mcrosporangiis aut conidiae: sporangiac globuiosac absens; sporophorum saepe vesiculatum]. Class 2. Zoomycelcs cl. nov. (diagnosis: parasites of animáis or protozoa; sporangiosporcs absent: in animaliis aut protozois parasitici, sporangiophora absens). Subclass 1. Entomycetidae subcl. nov. (diagnosis: endoparasites of invertebrates or protozoa; hyphac syncytial or subdivided by imperforate septa into coenocvtic segments; spores propulsive conidia; zygospores lateral A revised six-kingdom system o f life 247

T a b le 5. (cont.) to gamctangia: in animaliis aut protozois parasitiei; hyphae septatae aut non-septatae; conidia se praecipitantes) (orders Entomophoralcs, Zoopagalcs). Subclass 2. Pcdomycetidae subcl. nov. (diagnosis: symbionts of m andibulate arthropods, attached to their cuticle by foot-like holdfasts; imperforate plugged septa: in arthropodis mandibulatis parasiticac; in cuticula affixi haptero dilatato; septa imperforata). Superorder 1. Trichomycctalia supcrord. nov. (diagnosis: cndocommcnsals within the gui; spores are trichospores, formed by budding both asexually and following conjugation or arthrospores formed by hyphal fragm entation: intestinum incolcns; trichosporae aut arthrosporae inslructae) (orders Harpeflaies and ). Superorder 2. Pyxomycclalia superord. nov. (diagnosis: ectoparasites; differentiated male and female organs; female organ contains endospores and develops from three cells enclosed by a multicellular pseudoperithecium with an apical pore; male organ ail antheridium forming spcrmalia either endogenously or by exogenous budding: cctoparasilicae; pseudoperithecium endosporae continens; antheridium spermalia continens) (orders Laboulbeniales, e.g. Laboulbenia; Pyxidiophorales). P hylum 2 . Microsporidia Balbiani 1882 stat. nov. W eiser 1977 (diagnosis: obligate intracellular parasites of animals or rarely protozoa; vegetative cell walls, mitochondria and peroxisomes absent; spores with chitin walls and an extrusive polar tube through which the sporoplasm enters the new host cell). Class I. M inisporea Cavalier-Smith 1993 (diagnosis: polar tube with honeycomb outer layer) (e.g. Chytridiopsis, Buxlehudia, Hessia). Class 2. M icrosporea Levine & Corliss 1963 (diagnosis: polaroplast present: spores usually oval, rarely rod-shaped or pyriform). Subclass 1. Pleistophorea Cavalier-Smilh 1993 (diagnosis: multiply by plasmoiomv: one spore type: Pleistophorida Slempell 1906, e.g. Pleistophora , Amblyospora, Glugea, Eneephalitozoon). Subclass 2. Disporea Cavalier-Smith 1993 (diagnosis: multiply by binary ; disporogenic, i.e. two spore types, e.g..Nosema, Enterocytozoon , Spraguea, Caudospora). Subkingdom2 . N eom ycotasubking. nov. (diagnosis: usually with a dikaryotic phase; mciotic products form basidiosporcs or endospores or divide mitolically once to form ascospores; m itochondria and peroxisomes always present: cellulae binucleatae plerum que praesentes; endospora aut ascospora aut basidiospora instructa). Phylum 1. AscomycotaBerkeley 1857 slat nov. (diagnosis: mciotic products or their daughters form endospores by subdividing the cytoplasm by membrane, not by budding: sporae intracellulares). Subphylum 1. Hcmiascomycotina* Brcfcldt 1891 stat. nov. Ainsworth 1966 (diagnosis: ascocarp absent; usually ycasl-likc: sine ascocarpo; plerumque in forma fcrmenti). Class 1. Taphrinomvcetes cl. nov. (diagnosis: cell walls often lack chitin; mciotic products yield four endospores: muri saepe sine chitino endosporae quattuor, name introduced without Latin diagnosis by Cavalier- Smith, 1987 b) (e.g . Taphrina, Schizosaccharomyces, Prntomyces, Pneumocystis). Class 2. Geomyceles cl. nov. (diagnosis : with intracellular cyanobacteria! cndosymbionls (jVostoc sp .) ; hyphae coenocvtic when young; huge pale asexual spores form at hyphal lips; sex unknown: hyphac juvcncs sine seplis; cyanobacteria intra hyphis incolentcs) [Gcosiphonales ord. nov. (diagnosis as for Geomyceles) and sole family Gcosiphonaccac (Geostphon)]. Class 3. Endomycetes cl. nov. (diagnosis: budding ; chitin only in bud scars; ascogcnous hyphac absent; meiotic products four endospores formed by fusion of smooth cytoplasmic membranes around each nucleus separately; ascogcnous hyphac and dikaryotic phase absent: fcrmenti gcmmiparcs; chitinum in muro inter eellulis filiis, non in muri alteri; endosporae quattuor; validates the name introduced without Latin diagnosis bv Von Arx, 19 6 7 ). Subclass 1. Dipomvretidac subcl. nov. (diagnosis) (e.g. Dipodascopsis). Subclass 2. Saccharomycetidae de Bary 1866 slat. nov. ¡e.g. Saccharomyces, Candida, Endomyces). Subphvlum 2. Euascomycotina Engler 1897 slat. nov. Ainsworth 1966. (diagnosis: filamentous; chitin throughout the cell walls; ascus vesicle surrounds all four meiotic products; ascogenous hyphae with brief dikaryophase; ascocarp present in sexual forms: chitinum in muris omnibus; ascocarpa plerumque instructae). Class 1. Discomycctcs Fries 1836 stat. nov. Ainsworth 1966 (ascocarp an apothecium). Subclass 1. Calycomvcclidac subcl. nov. (diagnosis) Imoslly mazaedial , e.g. C alidum , Coniocybe). Subclass 2. Lccomycctidac subcl. nov. idiagnosis: asci inoperculale: asci sine opcrculis) (most fungi, e.g. Usnea, Lecanora, Peltigera, Xanthoria). Subclass 3. Pezomycetidae subcl. nov. (diagnosis: asci opcrculatc: asci opcrculati) (e.g. Peziza, Tuber, Morchella, Rhytisma).

[continued ovtrleaj 248 T. Cavalier-Smith

T a b l e 5 . icont.)

Class 2. Pyrenomycetes Fries 1821 stal. nov. Ainsworth 1966 (ascocarp a perithecium; mycelia septate when young). Subclass 1. V cnucom ycctidac subcl. nov. (diagnosis-, pyrcnocarpous lichen fungi: lungi lichcnis perithccia habentes) (e.g. Verrucaria). Subclass 2. Ostiomycctidac subcl. nov. (diagnosis: non lichen fungi: non lichenes) (e.g. pleurospora, Sordaria. Claviceps, A'ectria, Xylaria, Daldinia, some ). Class 3. Loculomycclcs cl. nov. (diagnosis: ascocarp not apothccial: ascocarpae non apotheciae). Subclass 1. Dendromycetidac subcl. nov. (diagnosis: lichcnizcd) ie.g. Arthonia, Lecanactis). Subclass 2. Loculoascomycetidae Lutlrell 1955 (ascocarps pscudothecial within a strom a; asci bilunicatc) (e.g. Dothidea, Pleospora, Ventana). Class 4. Plectomycetes Gwynnc-Vaughan 1922 stat. nov. Ainsworth 1966 (ascocarp a cleistolhecium ; asci unilunicatc) (e.g. Penecillium, Aspergillus, Eurotium, Erysipke; most fungi imperfecti). Phylum 2. Basidiomycotad e B a ry 1866em. auct. stat- nov. M oore 1980. [mciotic products form four exospores (ancestrally ballistosporcs) by budding from the surface of the cell; mciotic endospores absent; dikarya with clamp connections)]. Subphvlum 1. Septomycotina*î subphyl. nov. (diagnosis: uniporatc septa without doliporcs; basidia divided by transverse walls, but without long cpibasidia; ccntrosomes fiat: septa sine doliporis; muri transversi in basidiis pracbens, sine epibasidiis longis; centrosomac complanatae). Class 1. Seplomycetcs cl. nov. (diagnosis: ccntrosomes single layered: centrosomac monostromaticae: name introduced without Latin diagnosis by Cavalier-Smith, 19876; circumscription here narrowed by excluding the non-septate groups now placed in Ustomycctcs). Subclass 1. Sporidiomycclidac M oore 1980 stat. nov. (Erythrobasidiales, Sporidiales). Subclass 2. Uredomycetidae Brogniart 1824 stat. nov. (flat multilayered ccntrosomes) [rusts (Uredinales), ballistosporous and some other cxosporous yeasts (Seplobasidiales)]. Subphylum 2. Orthom ycotina subphyl. nov. (diagnosis: uniporate septa with doliporcs; ccntrosomes globular: septa doliporis m uri transversi in basidiis pracbcntcs; centrosomac globosae: name introduced without Latin diagnosis by Cavalier-Smith, 19876). Superclass 1. Hemibasidiomycclia Engler 1897 stat. nov. Class Ustomycctcs M oore 1980 (orders Ustilaginales, Tilletiales). Superclass 2. Hymenomycetia Fries 1821 stat. nov. cl cm. (typically with complex basidiocarp; ancestrally with an exposed hymcnium, but enclosed in polyphyletically derived subterranean fruiters; soma rarely reduced to phase). Class 1. Gelimvcetest cl. nov. (diagnosis: basidia divided by vertical walls ; gelatinous basidiocarp: muri longitudinali in basidiis praebens; corpus fructorum gelatinosum: name introduced without Latin diagnosis by Cavalier-Smith, 19876) (jelly fungi and related yeasts). Subclass 1. Tremellomycetidae Fries 1821 stat. nov. Wells 1994 (syn. class Exidiomycetes Moore 1994) subcl. nov. [diagnosis: muri longitudinali in basidiis praebens; corpus fructorum gelatinosus; basidia divided by vertical septa; saprobic jelly fungi (Tremellales) and related yeasts]. Subclass 2. Dacrymycetidac subcl. nov. [diagnosis: basidia furcate; basidia furcata (Dacrvmycclales). Subclass 3. Auromvcetidae basidia transversely septate with long cpibasidia (Auriculariales): basidia transversalilcr septata sed cpibasidiiis longis instructa]. Class 2. Homobasidiomycetes Patouillard 1900 (basidia not subdivided; basidiocarp usually non- gelatinous; no yeast phases). Subclass 1. Clavomycctidae Fries 1821 stat. nov. (e.g. Clavaria, Tulasnella). Subclass 2. Pileomycetldae Fries 1821 stat. nov. (e.g. Fomes, Agaricus, Caprinus. Boletus, Lycoperdon, Cyathus-, include gasteromycctcs, e.g. Phallus, Lycoperdon, Cyathus).

For morphological aspects of the phylogenetic basis of this classification sec Cavalier-Smilh (19876) ; for recent molecular sequence evidence on neomycotc phylogeny see Sugiyama & Nishida (1995). The nature of the kingdom Fungi and the origin of M icrosporidia are discussed in Cavalier-Smith (19986) Some evidence suggests that subphvlum Zygomycotina may be polyphvletic (Nagahama et al., 1995, Sugiyama, Nagahama & Nishida. 1996), so it may need to be split in future. Although the names Zygomycotina, Zygomycetes, and have been used widely for years, they appear not lo have been validly published, so I validate them here. * Probably paraphyletic. t Almost certainly paraphyletic. A revised six-kingdom system o f life 249 form a new subphylum Dictyomycotina, which VII. THE KINGDOM PLANTAE AND ITS FIVE therefore includes all fungi with well developed PHYLA Golgi dictyosomes, the ancestral state. Allomycetes are here grouped with the Zygomycotina (here The plant kingdom sensu Cavalier-Smith (1981a) ranked as an infraphylum and emended by removal comprises all organisms possessing plastids with of the Enteromycetes) in the new subphylum double envelopes that are free in the cytoplasm. Melanomycotina. The paraphyletic taxon Chytri- Unlike chromists and protozoa, all plants are diomycotina, which has been long used despite never obligately dependent on plastids and have never lost being validly published (for higher taxa mycologists them even when they have become secondarily non- have often ignored the rule of the International photosynthetic : this is probably because they are Code of Botanical Nomenclature that requires Latin essential for the synthesis of fatty acids, starch and diagnoses for the valid publication of new names) is certain amino acids (Cavalier-Smith, 1993e). As the now abandoned. evidence for the single origin of the chioroplast from The remaining Trichomycetes ( and a cyanobacterium and for the monophyly of Plantae Asellariales) are reduced in rank to a superorder has recently been discussed in detail (Cavalier- (Trichomycetalia : there is no standard mycological Smith, 1995 a), and the circumscription of Plantae suffix for a superorder), and placed in a new has remained unchanged since Cavalier-Smith zygomycote class exclusively m ade up of parasites of (1 9 8 1 a ), I shall not discuss them further here, except animals or protozoa, which I call Zoomycetes, and to stress that it is not the presence but the divide into two subclasses. One subclass (Pedomyce- morphology and location of the plastids, together tidae) includes the Trichomycetalia and a second with their obligate nature, that define the kingdom. new superoder (Pyxomycetalia) created for the Table 6 summarizes the classification of the king­ Laboulbeniales and Pyxidiophorales (Blackwell, dom ; from rRNA trees it is clear that the three major 1994), ectoparasites of arthropods formerly regarded lineages, , Glaucophyta and Rhodo­ as ascomycetes because of the superficial resemblance phyta diverged almost simultaneously, so closely of their female organs to perithecia. Both Tricho­ following the origin of chloroplasts that it is hard to mycetalia and Pyxomycetalia attach to their host determine their correct branching order or even to cuticle by similar foot-like holdfasts (whence the corroborate the monophyly of the kingdom by gross name Pedomycetidae) and have similar imperforate sequence similarity; but there is other molecular plugged septa and a body form of discrete cells not evidence for the monophyly of Plantae in the present indefinite hyphae. The other new zoomycete subclass sense (Ragan & Gutell, 1995). The monophyly of is the Entomycetidae, comprising the orders Ento- Plantae has been questioned on the basis of RNA mophorales and , tissue endoparasites of polymerase sequence trees (Stiller & Haii, 1997), but protozoa or invertebrates, which have usually been there are too few taxa yet on these trees to have high included in the Zygomycetes. Two other new classes confidence in their branching order. (the saprotrophic Bolomycetes and the endo- mycorrhizal Glomomycetes) have also been tra­ (1) New red algal subphyla ditionally placed in the Zygomycetes, which are here restricted to Zygomycotina with stalked sporophores Currently are divided into two classes: the bearing passively dispersed aerial spores (e.g. the paraphyletic and holophyletic Flori­ well-known ). The present four zygo­ deophyceae. Bangiophyceae are ultrastructurally so mycote classes are all very much more homogeneous diverse that I divide them into two classes: Bangio­ than the traditional Zygomycetes and Tricho­ p h y c e a e sensu stricto (Bangiales, Rhodochaetales) mycetes. Though there are no sequences yet avail­ which have pit connections and able for trichomycetes, those for other zygomycetes (Porphyridiales, Cyanidiales, Compsopogonales) show their great heterogeneity and justify their which do not. In having pit connections, and on subdivision into several classes (Nagahama el a i , rRNA trees, Bangiales are closer to Florideophyceae 1995). The present system, however, emphasizes (therefore I group them as the new subphylum homogeneity of morphology and does not slavishly Macrorhodophtina) than to the Rhodellophyceae, follow rRNA trees, since in Fungi these are highly which I place alone in the new subphylum Rhodel- unclocklikc, with numerous unusually long lophytina. The Porphyridiales are themselves very branches, and so must in some respects be very diverse and should probably be subdivided into misleading. more than one order. 250 T. Cavalier-Smith

T a b le 6. Classification o f the kingdom Plantae and its five phyla

Subkingdom 1. Biliphyta*Cavalier-Smith 1981. Infrakingdom 1. Glaucophytainfrak. nov. (diagnosis: pcptidoglycan in plastid envelope: plastidae peptidoglycanum instructae) (glaueophytes). Phylum Glaucophyta Skuja 1954 (unnecessary syn. Glaucocyslophyla Kies & Krcmcr 1986) (e.g. Cyanophora). Infrakingdom 2. Rhodophytainfrak. nov. (diagnosis: plastid envelope lacks peptidoglycan: sine peplidoglycano). Phylum Rhodophyta W ettstein 1922 (red algae). Subphylum 1. Rhodellophytina* subphyl. nov. (diagnosis: unicellular, e.g. Porphyridium, or undifferentiated simple or branched uniscriate filaments with a basal disc, e.g. Stylonem a ; pit connections absent: ccllulae unicae aut filamentae simplices uniseriatae cum disco basalo: sole class: Rhodellophyceae cl. nov. diagnosis as for subphylum). Subphvlum 2. M acrorhodophytina subphyl. nov. (diagnosis: multicellular with extensive cell differentiation and complex life histories; pseudoparenchymatous or multiseriate filaments or fiat sheets of cells; pit connections usually present: thallus multicelllularis; saepe in forma pseudoparenchyma aut filamentae multiseriatae) (classes: Bangiophyceae em. ; Floridiophyceae). Subkingdom2 . Viridiplantae Cavalier-Smith 1981 (green plants). Infrakingdom 1. ChlorophytaCavalier-Smith 1993. Phylum Chlorophyta auct. Subphylum 1. Chlorophytina subphyl. nov. (diagnosis: cytokinesis without a phragmoplast; multilayered structure (MLS) ciliary roots absent except in . Pterosperma. Halosphaera sine phragmoplastis) (chlorophyte green algae). Infraphylum 1. Prasinophvtaef infraphyl. nov. (diagnosis: usually scaly flagellates with persistent telophase spindle; without phycoplast; M g 2, 4 divinyl porphyrin often used as a photosynthelic pigment: ccllulae cilio aut ciliis instructa, usitate squam atae; fusus persistens in telophaso; sine phycoplasto) [classes M icromonadophvceae M attox & Stewart 1984 em. with cruciate ciliary roots: orders Mamiellales M ocstrup 1984 cm., em., Mesostigmatales ord. nov. (diagnosis: biflagellates; cruciate ciliary roots with one M LS; tubular ciliary hairs absent: cilia dua; radices mlcrotubulorum in forma crucis cum structura laminarum m ullarum ; cilia sine pilis tubulalis; formerly included in Pyramimonadales); Ncphrophyccac Cavalier-Smith 1993, with asymmetric m icrotubular roots, e.g. Nephroselmis, Pseudoscourfieldia). Infraphylum 2. Tclraphytac infraphyl. nov. (diagnosis: cruciate m icrolubular roots without MLS; telophase spindle collapses; cytokinesis involves a phycoplast: radices m icrotubulorum in forma crucis; sine structura laminis multis) (e.g. Ulva, Chlamydomonas, Chlorella', type genus Tetraselmis). Subphylum 2. Phragm ophvtina subphyl. nov. (diagnosis: vegetative cells non-motile; zoospores or sperm with asymmetric ciliary roots typically with M LS; cytokinesis usually involves a phragmoplast) (lo avoid ambiguity it is best to call these plants phragmophytcs informally, and to reserve the term charophyte strictly for the infraphylum Charophytac). Infraphylum I. Charophytae Engler 1887 stat. nov. (haploid multicellular macrophytcs with main axis corticated with whorls of lateral branches; with oogonia and anthcridia surrounded by sterile cells) (class Charophyceaesensu stricto: order Charales: stoneworts - the traditional charophytes). Infraphylum 2. Rudophytae infraphyl. nov. [diagnosis: antheridia absent or without investing sterile cells: anthcridia non cellulis sterilis circumcincta: (descriptive name: suggested vernacular term: rudophytes, meaning simple - Latin rudis - plants)][classes Eophyceae cl. nov. (diagnosis: with biciliate zoospores or sperm: zoosporac aut gatnetae musculinae ciliis instructae: the name cophytc is chosen because the Chaetophycidae may be ancestral to land plants: Graham , 1993) subclasses Stichophvcidae subcl. nov. (diagnosis: sarcinoid or unbranchcd filaments: cells lack sheathed hairs: thallus non ramosus; cellulae sine sclis vaginalis; descriptive name to indicate that the most complex forms have only a single row - stichus - of cells) (orders Chlorokybales. Klebsormidiales) ; Chaetophycidae cl. nov. (diagnosis: many cells have protective sheathed hairs (chactac) : cellulae selis vaginatis munilac) branched filaments or parenchymatous with sheathed hairs () ; Conjugophvceae auct. (all cells, including gametes, lack cilia) (Desmidlales, Zygnematales)]. Infrakingdom 2. CormophytaE n d lic h e r 1836(syn. Embryophyla auct.) stat. nov. Phylum 1. Bryophyta Braun 1864 em. Eichler 1883 (, liverworts, ). Subphylum 1. Hcpaticae auct. slat. nov. (liverworts, e.g. Riccia, Marchantía, Lophocolea). Subphylum 2. Anthocerotac auct. stat. nov. (e.g. Anthoceros). Subphylum 2. Musei Linnaeus 1753 stal. nov. Infraphylum 1. Sphagncae auct. slat. nov. (leaves with large porous dead cells as well as living ones; Sphagnum ). A revised six-kingdom system o f life 251

T a b ic 6. (cont.)

Infraphylum 2. Brvatac infraphyl. nov. (diagnosis: cells in leaves all alive with chloroplasts: cellulae omnis in laminis chloroplastis inslructae) (e.g. Andreaea, Bryum, Funana ). Phylum 2. Tracheophyta phyl. nov. (diagnosis: diploid phase with xvlem and phloem: facies diploida xylem et phloem instructa: name introduced without Latin diagnosis by Sinnolt, 1935). Subphylum I. Pteridophytina! Eichler 1883 stal. nov. (pteridophytes). Infrapliyluin 1. Psiiophytae infraplil. nov. (sine radicibus; without roots, e.g. Psilophytan , P silolum ). Infraphylum 2. Lvcophytae auct. slat. nov. (e.g. Ltpidodendron , lycopods, Selaginella, ¡roetes). Infraphylum 3. Sphcnophytae auct. slat. nov. (e.g. Sphenophyllum, horsetails). Infraphylum 3. Filiccs* Linnaeus 1753 slat. nov. (). Subphylum 2. Spermatophytina auct. stat. nov. (seed plants). Infraphylum 1. Gymnospcrmae auct. (‘seed ferns', , , gnetophylcs). Infraphylum 2. Angiospcrmac auct. (dicol and monocoi flowering plants).

(2) New chlorophyte infraphyla VIII. CHROMISTA, THE THIRD BOTANICAL KINGDOM, AND ITS FIVE PHYLA It has long been accepted that the ancestral green plants are the scaly prasinophyte algae (Mattox & Stewart, 1984), and that there was a basic bi­ Most chromists are algae with chloroplasts con­ furcation within green algae between the Chtoro- taining chlorophyllsa a n d c which are located not in phytina and Charophytina. Because cormophytes the cytosol (as in plants or in protozoan algae) but evolved from a charophytine algae, Charophytina within the lumen of the rough endoplasmic reticu­ have often been treated as a separate phylum lum. In addition to their double chioroplast envel­ (Jeffrey, 1971; Cavalier-Smith, 1993c, 1995a) to opes, chromistan plastids are surrounded by an enable and Cormophyta to be grouped additional smooth membrane, the periplastid mem­ together as an infrakingdom . It now brane (Cavalier-Smith, 1989 b ). All chromistan algae appears that the prasinophyteM esostigm a is cladis- are evolutionary chimaeras between a eukaryotic tically closer to the traditional charophytes than to host and a eukaryotic (probably red algal) symbiont; the other prasinophytes (Melkonian, Marin & the periplastid mem brane is the relic of the red algal Surek, 1995). Because of this and the fact that it is plasma membrane. The unique location of plastid the only prasinophyte with a multilayed structure, it plus periplastid membrane within the rough en­ was recently transferred from the Chlorophyta to the doplasmic reticulum arose by the fusion of the Charophyta (Cavalier-Smith, 1995a). However, this former phagosomal membrane that first enclosed the further diminishes the phenotypic difference be­ endosymbiont with the nuclear envelope (Whatley, tween these two major green algal taxa; therefore John & Whatley, 1979; Cavalier- Smith, 1982, though the recognition of two green algal phyla is 1986 a). The evidence for the monophyly of the cladistically attractive it creates a boundary at a Chromista is discussed in detail elsewhere (Cavalier- point where the phenotypic difference seems too Smith, 1995a). Recent molecular evidence (Cava­ slender to justify separate phyla. Therefore I revert lier-Smith, Allsopp & Chao 1994a; Cavalier-Smith to the more traditional simpler system with a single & Chao, 1997a) makes it clear that chlorarachnean green algal phylum Chlorophyta divided into two algae, originally excluded from Chromista but new subphyla. Streptophyta remains as a clade temporarily transferred into this kingdom because of n a m e only. misleading rRNA trees (Cavalier-Smith 1993 a, As before (Cavalier-Smith, 1993r), Cblorophytinn 1994) are actually Protozoa, as considered previously are divided into two major taxa (now ranked as (Cavalier-Smith 1986a). infraphyla) according to their cell division mecha­ Chromist monophyly remains controversial for nism: Prasinophytae (biciliate or uniciliate with two reasons; one is that the three major chromist persistent spindle) and Tetraphytae with collapsing groups (Cryptista, Heterokonta and Haptophyta; spindle and phycoplast. Charophytina are also sub­ see Table 7) diverged from each other almost divided into two infraphyla: Charophytae and simultaneously with the three major plant lineages, Coleophytae. and the six lineages can appear to diverge in almost 252 T. Cavalier-Smith

T a b le 7. Classification o f the kingdom Chromista and its five phyla

Subkingdom 1. CryptistaCavalier-Smith 1989- Phylum Cryptophyta Cavalier-Smilh 1986 {e.g.. ). Subkingdom2. ChromobiotaCavalier-Smith 1991. Infrakingdom 2 . Heterokonta Cavalier-Smith 1986 stat- nov. 1995 em. Superphylum 1. superphvi. nov. (Diagnosis as for phylum Sagenista Cavalier-Smilh 19956 p. 1014). Phylum Sagenista Cavalier-Smilh 1995. Subphylum 1. Bicoecia Cavalier-Smith 1989 [orders Bicoccalcs Grasse & Deilandre 1952 (Bicoeca), Anoccalcs Cavalier-Smith 1997 (e.g. )-, Pirsoniales ord. nov. [diagnosis: syncytial non-motile trophic phase that feeds on diatoms; with trophosome inside and auxosome outside the frustule; zoospores with non-iubular hairs on cell body: diatomas dévorant: in statu pabulatorio cellulae nuclei plurcs instructae; trophosoma sine nucleis intra fruslulo, auxosoma m ultinucleata extra frustulo; zoosporae pili non-tubulati vestitae) (sole family Pirsoniaceae fam. nov. diagnosis as for order Pirsoniales: type genus Pirsonia]. Subphylum 2. Labvrinthista Cavalier-Smith 1986 stat. nov. 1989 (e.g. , Thraustochytrium). Supcrphylum 2. Gyrista superphyl. nov. (diagnosis: ciliary transition region commonly with a helix - or a double helix/ring system: regio transitoria ciliorum plerumque hcliccm aut helices praebens). Phylum 1. Ochrophyta* Cavallcr-Smith 1986 (as Ochrista) stat. nov. 1995. Subphylum 1. Phaeista Cavalier-Smith 1995. Infraphylum 1. Hypogyrisla Cavalier-Smith 1995 (e.g. pcdineliids, silicoflagcllates, pelagophytes). Infraphylum 2. Chrysista Cavalier-Smith 1986 stat. nov. 1995 (superclasses Phaeistia Cavalier-Smith 1995, e.g. , xanthophytcs,Chrysomeris; Limnistia Cavalier-Smilh 1996, e.g. chrysophytcs, custigs, Oikomonas, raphidophytcs). Subphylum 2. Diatomeae Dumortier 1821 slat. nov. Cavalier-Smilh 1995 (centric and pennate diatoms). Phylum 2. Bigyra phyl. nov. (diagnosis: ciliary transition region with double helices or concertina-like rings; without plastids: regio transitoria ciliorum annuli in forma concerlinae praebens: sine plastidis). Subphylum 1. Bigyromonada subphyl. nov. (diagnosis: biciliale free-living bacterivorous phagotrophs without cell walls; relronemes on anterior cilium; double ciliary transition helix: cilia dua; murus absens; mastigonemae tubulatac in cilium anterius; pabulum cellulae prokaryotac est; regio transitoria ciliorum annuli in forma conccrtinac praebens) (Developayella Tong 1995). Subphylum 2. Pseudofungi Cavalier-Smith 1986 em. 1989 (oomycetes, hyphochytrids). Subphylum 3. O palinata W enyon 1926 slat. nov. cm. Cavalier-Smith 1993, 1997 (diagnosis: retronemes absent from cilia; gut commensals without peroxisomes: sine mastigoncmis tubularis in ciliis) { , Karotomorpha, opalinids). Infrakingdom 2 . Haptophyta Cavalier-Smith 1995. Phylum Haptophyta Cavalier-Smith 1986 (e.g. Pavlova, ).

For a more detailed classification of Ochrophyta and Haptophyta see Cavalier-Smith & Chao (19966) and Cavalier-Smith et al. (1996¿) respectively. * Probably paraphyletic. I Almost certainly paraphyletic.

any order on different molecular trees; the other is that rapidly diverged into the three major lineages that all three of the m ajor lineages contain some non- (Cavalier-Smith, 1982, 1986a). Given our present photosynthetic species, and it has been hard to knowledge of chromist phylogeny, one has to establish whether these are primarily or secondarily postulate a minimum of six independent losses of photosynthetic. M any authors, following Margulis chloroplasts within the kingdom. (1970) and W hatley et al. (1979), have assumed that all are primarily non-photosynthetic and that there (1) Multiple losses of chloroplasts by were three (or even more) independent implanta­ c h r o m is t s tions of eukaryotic algae into unrelated protozoan hosts. In contrast, I have argued that all are F.videncefor chioroplast loss by chromists is growing; secondarily non-photosynthetic and originated by it is strongest for the Heterokonta. We now have several independent losses of chloroplasts and that strong molecular phylogenetic evidence for three all chromists had a common photophagotrophic losses (two in pedinellids: Cavalier-Smith, Chao & ancestor-the first eukaryote-eukaryote chimaera Allsopp, 1995; Cavalier-Smith & Chao, 19966; and A revised six-kingdom system o f life 253 o n e in O ikom onas: Cavalier-Smith et a l., 19 9 6 ¿i lack tubular hairs, in the subphylum Opalinata within the predominantly algal phylum Ochrophyta (Cavalier-Smith, 1997a). For reasons discussed else­ (Cavalier-Smith & Chao, 1996Ä). We have weaker where, I have recently left this taxon outside the evidence for a third loss in the ancestor of the kingdom Protozoa (Cavalier-Smith, 1997a). I now Pseudofungi (which comprise the classes Oomycetes place this subphylum within the Chromista in the and Hyphochytrea, traditionally thought of as infrakingdom Heterokonta. This change is made as fungi), which may have evolved from an ochrophyte a result of the recent discovery of a distinctive new algal ancestor by plastid loss (Cavalier-Smith et at., type ofheierotrophic heterokont,Developayella elegans 1996 Recent maximum likelihood trees for rRNA (Tong, 1995), and a reappraisal of the evolution of also suggest that Goniomonas, the sole aplastidic genus the ciliary transition region and cell surface in in the phylum Cryptophyta, may also have arisen by ch ro m ists. plastid loss (Cavalier-Smith et a!., 1996c). Thus, we Developayella elegans resembles both Pseudofungi have strong evidence from the internal phylogeny of and O palinata in having a double helix distal to the the Heterokonta and Cryptista for three, and some transverse plate in the ciliary transitional region, evidence for five, of the six plastid losses required by which somewhat resembles a concertina in longi­ the theory of a photophagotrophic latest common tudinal section. The structure of the ciliary transition ancestor for chromists. We lack such internal region has proved to be an excellent phylogenetic evidence only for one established chromist group, marker in the past; for example the nine-fold star the phylum Sagenista (bicoecids, thraustochytrids that characterizes the subkingdom Viridiplantae and labyrinthulids), which diverges from other (M antón, 1965; Cavalier-Smith, 1981a). Recently, heterokonts near the base of the heterokont clade on I have given it considerable weight in revising the rRNA trees, before the earliest divergences within higher level classification of the heterokont algal the mostly photosynthetic Ochrophyta. Since mol­ phylum Ochrophyta (earlier spelled Ochrista) ecular trees (contrary to what has been implied by {Cavalier-Smith, 1995a). I have divided the ochro­ some authors) do not refute the quite strong phyte subphylum Phaeista into two infraphyla: morphological and chemical evidence for the mono- Hypogyrista, characterized by a short single ciliary phyly of Chromobiota (discussed in detail by transitional helix located below (proximal to) the Cavalier-Smith, 1994; Cavalier-Smith e ta l., 19 9 6 ¿Ti, transition plate and Chrysista characterized by a it is much more parsimonious to postulate that longer single helix above the transition plate. These Sagenista also have lost chloroplasts than to suggest differences are congruent with other morphological that Ochrophyta and Haptophyta acquired indis­ differences and with molecular phylogeny (Cavalier- tinguishable fucoxanthin-containing chloroplasts in­ Smith & Chao, 1996¿). It therefore seems likely that dependently, as some authors (e.g. Leipeel a l , 1994; the double transitional helix is also a strong Bhattacharya & Medlin, 1995) assume. phylogenetic character. Accordingly, I placeD e ­ velopayella elegans, Pseudofungi and Opalinata to­ gether to form a new heterotrophic heterokont phylum, the Bigyra: the name refers to the two (2) Transfer of Opalinata to Chromista: the helices or sets of rings (which interpretation is correct new heterokont phylum Bigyra is not entirely clear) in the ciliary transition region W hen the kingdom Chromista was first established (from Latin bí-, meaning ‘ twice’, and gyrus, m e a n in g (Cavalier-Smith, 1981

absence of such cortical microtubules in the ancestor konts or cryptophytes that have directly lost retro­ of gregarines would explain why their folds are nemes, though such loss is not impossible if an instead supported by actin microfilaments. Likewise alternative mode of feeding and/or taxis evolved at the attachm ent of retronemes to ciliary microtubules the same time. The remote possibility that foramini­ was a preadaptation to their subsequent attachment fera are heterokonts (Patterson, 1989) is now to cortical microtubules. excluded by rRNA phylogeny (Pawlowski et al., The loss of somatonemes would in principle be 1994, 1996; Wray et a t., 1995). are, easier than the loss of retronemes as it would not however, known to be able to lose the haptonema reverse the direction of ciliary hydrodynamic flow. entirely and some have lost the patelliform scales Thus the conversion of retronemes to the somato­ found in most (Cavalier-Smith et n em es o f Proteromonas facilitated their subsequent a l., 1996¿) ; moreover a few are non-photosynthetic total loss. (Marchant & Thomsen, 1994). A non-photo- After the present paper was submitted rRNA synthetic that had lost both the hapto­ sequence evidence was published (Silberman et al., nem a and scales could easily be wrongly classified as 1996) showing that Proteromonas lacerlae-viridis is a protozoan. Although there is no positive reason to specifically related to the heterokonts, and is there­ think that any such misclassified haptophytes exist, fore not an outgroup for chromists as a whole. This the case of the haptonema-less haptophyteReticulo­ adds further support to the arguments given above sphaera japonensis that was misclassified as a hetero­ for the view that retronemes were secondarily kont (Grell, Heini & Schüller, 1990) before mol­ transferred from the cilia to the cell surface of the ecular evidence for its true nature was found ancestral opalinate, and for transferring the Opali­ (Cavalier-Smith et a i , 1996if) should alert one to the nata as a whole into the Heterokonta; however, possibility that some zooflagellates presently thought a lth o u g h Proteromonas clearly clusters with the un­ of as protozoa might actually be drastically altered doubted heterokonts, the published tree does not haptophytes. actually group it with the Pseudofungi as would be Though the loss of retronemes must be very rare, expected if the Bigyra are holophyletic, as postulated it is, in principle, much easier for chromists to lose here. Since however the bootstrap values for the their flagella totally. This has long been known to branching order at the base of the Heterokonta are have occurred in pennate diatoms and probably in low', and since the deepest branches are rather long, Chrysamoeba, and has recently been demonstrated for it is possible that some or even all of them are placed Pelagococcus a n d Aureococcus, though as these are all too low in the tree, as may be often be true for the photosynthetic they have not been mistaken for Pseudofungi themselves (Cavalier-Smith et al., protozoa. If photosynthesis were also lost and the 19966). Additional evidence is therefore needed to had no other characters identifying it as a test the monophyly of the Bigyra. Silbermanet al. chromist it could be wrongly treated as a non- (1996) also show that the non-ciliated gut symbiont flagellate protozoan, just as was done for . Blastocystis is related to Proteromonas , w h ich w as It is conceivable that at least a few others of the entirely unexpected. This means that despite the numerous parasitic protists of uncertain taxonomic absence of cilia Blastocystis should be placed in the position listed by Patterson (1994) might actually be Opalinata, and that it is not a fungus or protozoan chromists. Several hundred protist genera (those as previously supposed. As it differs from all other listed in Table 5 of Patterson, 1994, minus a few like Opalinata in the absence of cilia I place it in a new- Blastocystis, Dermocystidium a n d T rim a stix , which have class, Blastocystea (diagnosis: sine ciliis - w ith o u t now been put in phyla) are so poorly studied that cilia). Blastocystis is the first chromist known to they cannot with confidence be placed in any of the parasitize humans. These new data are discussed in kingdoms or phyla of the present system, and mostly more detail elsew'here (Cavalier-Smith, 19976). not yet been placed in any suprageneric taxon. Though the vast majority of these genera will probably turn out to belong somewhere in the (6) Are there other unidentified chrom ists infrakingdom Sarcomastigota of the subkingdom lurking within the kingdom Protozoa? Neozoa of the kingdom Protozoa (most I suspect in For the reasons discussed above, the direct loss of the Cercozoa), a small minority might turn out to be retronemes by phagotrophic heterokonts is so im­ chromists, once they are studied by molecular probable that it is unlikely that any of the well phylogenetics. characterized zooflagellate protozoa are really hetero­ Patterson (1989) reasonably suggested that D ipla- 258 T. Cavalier-Smith phrys a n d Sorodiplophtys may actually be non-fla- Opalinata from Protozoa to Chromista; of Micro­ gellate thraustochytrids, but molecular evidence is sporidia from Protozoa to Fungi; and of Myxozoa required to confirm or refute this. One other and Mesozoa from Protozoa to Animalia. One can o rg a n ism ( limacisporum) in itia lly expect the boundaries of the six kingdoms to be even treated as a non-flagellate thraustochytrid (Raghu- more stable in the future. W hile there remains a need Kumar, 1987) has recently been shown by rRNA for a yet more thorough testing of the monophyly of phylogeny to be a non-flagellate choanozoan proto­ Plantae and Chromista, the present six-kingdom zoan instead (Cavalier-Smith & Allsopp, 1996). system is phylogenetically and taxonomically Earlier Davidson (1982), Patterson & Fenchel sounder, for the reasons discussed previously (1985) and Patterson (1986) proposed that actino- (Cavalier-Smith, 1986a, 1993a), than the five- phryid heliozoa were non-flagellate heterokonts; kingdom system found in its numerous mutually though Patterson (1994) appears no longer to contradictory variants in most textbooks. It is also support this unconvincing view, the true position of distinctly simpler, and so practically more con­ actinophryids still needs to be checked by rRNA venient and easier for beginning students and general phylogeny. Patterson’s (1994) informal unranked users to comprehend, than the phylogenetically group ‘stramenopiles’ is identical in phylogenetic congruent eight-kingdom system that I advocated concept to the infrakingdom Heterokonta. Apart from 1987 to 1995. I hope that it will be widely from the exclusion of Blastocystis, which prior to the a d o p te d . rRNA evidence mentioned above showed no obvious The major advances over the past 15 years have signs of its heterokont affinities, the inclusion of been in the phylogenetically sounder definition of Reticulosphaera, which is now known to be an error, the phyla, subphyla and infraphyla, involving many a n d C om m ation, and the uncertainty about the new creations, subdivisions and mergers of major position ofD iplophrys , Patterson’s (1994) '.strameno­ taxa, mainly in the bacteria, protozoa and hetero­ piles’ is also identical in composition to the infra- kont Chromista, and in the definition of sub­ kingdom Heterokonta as revised here. I do not kingdoms and infrakingdoms of all kingdoms. There accept the inclusion of Comm ation in Heterokonta is probably still significant scope for further improve­ (Thomsen & Larsen, 1993 ; Patterson, 1994), as I am ments in these respects, especially amongst the unconvinced that these protists have bipartite or neozoan protozoa, and a need for more detailed tripartite retronemes; I have placed Commation testing of the recent changes, but it is likely that the instead in its own order within the protozoan phylum spate of new creations of higher level taxa that has Neomonada in the class Kinetomonadea (Cavalier- accompanied the recent extension of electron mi­ Smith, 1997a). In view of the ultrastructural croscopy and molecular phylogeny into previously similarities with Heliomonadida, I have grouped unexplored territory will be much reduced in the Heliomonadida and Commatiida together as the ensuing decades, and we can look forward fairly soon subclass Ramicristia (Cavalier-Smith, 1997a). The to a period of consolidation and relative stability in reasons for not using the unnecessary new synonym the classification and nomenclature of the major in preference to the classical term groups of life. heterokont were expounded previously (Cavalier- Smith, 1993a). In sum, though the Chromista may, in future, X. CONCLUSIONS need to be augmented slightly by the addition of a few misplaced heterotrophs, it is unlikely that the 1. An outline classification of a revised six-kingdom boundary between Chromista and Protozoa will system of life is presented, down to the level of change substantially, and even possible that both infraphylum. Intermediate very high level categories kingdoms will be entirely stable in circumscription in (superkingdom, subkingdom, branch, infrakingdom, the future. superphylum, subphylum and infraphylum) are extensively used to avoid splitting organisms into an excessive number of phyla (only 60 being recog­ IX . E N V O I nized) and kingdoms, and to achieve a balanced system without unwarranted lumping. In the present six-kingdom system the only changes 2. The circumscription and high level classifi­ in the circumscription of the six kingdoms from that cation of the zoological kingdoms Protozoa and of Cavalier-Smith (1983a) are the transfer of Animalia are modified in the light of recent A revised six-kingdom system o f life 259 molecular phylogenetic evidence that the protist the recently discovered heterotrophic heterokont Myxozoa are actually Animalia, not Protozoa, and Developayella elegans (placed here in the new sub­ that mesozoans are bilaterians. Mesozoa are re­ phylum Bigyromonada) to form a new purely moved from the kingdom Protozoa and placed as a heterotrophic botanical phylum, Bigyra. Bigyra are new infrakingdom within the subkingdom Bilateria defined as heterotrophs with a double ciliary of the Animalia, which therefore are now subdivided transitional helix. This new grouping makes it into three subkingdoms: Radiata (phyla Porifera, necessary to abandon the phylum name Opalozoa, Cnidaria, Placozoa, Ctenophora), Myxozoa, and in which Opalinata were previously placed. A Bilateria (bilateral animals: all other phyla). detailed evolutionary theory is presented to account 3. Microsporidia other than the metchnikovellids for the loss of ciliary retronemes in Opalinata as a are transferred from the kingdom Protozoa to the consequence of their evolution of gut commensalism. kingdom Fungi. Recent phylogenetic evidence for multiple chloro- 4. I argue that the need for a simple and general plast losses within the kingdom Chromista strength­ classification of the living world is now best m et by ens the view that the ancestral chromist was a placing Archezoa as a subkingdom within the photophagotroph chat evolved by a single symbio- Protozoa, as in my 1983 six-kingdom system, rather genetic event involving a phagotrophic biciliate host than as a separate kingdom as I advocated from and a red algal endosymbiont; but the monophyly of 1987 onwards. I group the 13 currently recognized the Chromista still needs to be tested more. protozoan phyla into two subkingdoms, Archezoa 7. No changes are made to the circumscription of and Neozoa, and four neozoan infrakingdoms. The the botanical kingdom Plantae or the kingdom reasons for these changes are discussed in detail in Bacteria, which have both been stable since relation to the principles of megasystematics, here Cavalier-Smith (1981

A m aral Z ettler, I,., Sogin, M . I- &. C aron, D. A. (1997). C a n n i n g , E. U (1990). Phylum Microspora. In: Handbook of Phylogenetic relationships between the and the Proloclista (ed. I. Margulis, J. O. Corliss, M. Melkonian and Polycvstinea: a molecular perspective on Haeckel's Radiolaria. D. J. Chapman), pp. 53-72. Jones át Bartlett, Boston.

Proceedings of the National Academy of Sciences, USA 94, C a r l i l e , M .J . & W a t k i n s o n , S. C . ( 1 9 9 4 ) . The Fungi 11411-11416. Academic Press, London.

A s h l o c k , P.D . ¡1971). Monophyly and associated terms. C a v a l i e r -S m i t h , T. (1978). The evolutionary origin and Systematic Zoology 20, 6 3 - 6 9 . phylogeny of microtubules, mitotic spindles and eukaryote B aldauf, S. I,. & P a l m e r , J . D . ( 1 9 9 3 ) . Animals and f u n g i a r e flagella. BioSystems 10, 93-113. each other’s closest relatives, congruent evidence from C a v a l i e r -S m i t h , T. (1980). Cell compartmentation and the m ultiple proteins. Proceedings o f the National Academy o f Sciences, origin of eukaryotic membranous organelles. In Endocyto - USA 90, 11558-11562 biology: Endosymbiosis and Cell Biology (cd. W. Schwemmler and B a l d a o f , S. I.., P a l m e r , J . D. & D oolittle, W . P. ( 1 9 9 4 ) . T h e H. E. A. Schenk), pp. 831-916. De Gruytcr, Berlin. root of the universal tree and the origin of eukaryotes based on C a v a l i e r -S m i t h , T . (1981a). Eukaryote kingdoms, seven or clogatlon factor phylogeny.Proceedings o f the .S'ational Academy n in e ? BioSystems 14, 461 481. o f Sciences, USA 93, 7 7 4 9 - 7 7 5 4 . C a v a l i e r -S m i t h , T . (1981 b). The origin and early evolution of B a l d a u f , S. I,. & D oolittle, V\’. F. (1997). Origin and the eukaryotic cell. In Molecular and Cellular Aspects o f Microbial evolution of the slime moulds. Proceedings o f the National Evolution. Society for General Microbiology Symposium 32 Academy o f Sciences, USA 9 4 , 12007-12012. (cd. M .J. Carlile, J. F. Collins and B. F.. B. Moseley), pp. B haitacharya, D . & M edlin, L. 1 1995 ). The phylogeny of 33-84. Cambridge University Press. plastids, a review based on comparisons of small-subunit C a v a u e r -S m i t h , T. (1982). The origins of plastids. Biological ribosomal RNA coding regions. Journal of 31, Journal of the Linnean Society 17, 289-306. 4 8 9 -4 9 8 . C a v a l i e r -S m i t h , T . ( 1 9 8 3 a ) . A 6-kingdom classification and a B a r n s , S. M ., D e l w i c h e , C . F ., P a l m e r , J . D . & P a c e , N . R . unified phylogeny. In Endocylobiology II (cd. H. E. A. Schenk (1996). Perspectives on archcal diversity, tbermophily and and W. Schwemmler), pp. 1027-1034. De Gruyler, Berlin. monophyly from environmental rRNA sequences. Proceedings C a v a l i e r -S m i t h , T. (1983é). Endosymbiolic origin of the o f the National Academy : f Sciences, USA 93, 9 1 8 8 -9 1 9 3 . mitochondrial envelope. In Endocytobiology II (ed. W . B lackw ell, M . ( 1 9 9 4 ) . M in u te mycological mysteries: the Schwemmler and H. E. A. Schenk), pp. 265-279. De Gruyter. influence of arthropods on the of fungi. M ycologia 8 6 , B erlin. 1 - 1 7 . C a v a l i e r -S m i t h , T. (1986a). The kingdom Chromista, origin B l o b e l , G. (1980). Intracellular protein topogcnesis. Proceedings and systematics. In Progress in Phycological Research (cd. F. K. of the National Academy o f Sciences, USA 77, 14 9 6 -1 5 0 0 . Round and D. J. Chapman), vol. 4, pp. 309-347. Biopress, B r a d l e y , P . J . , L a h t i , C . J . , P l u e m p e r , L . & Johnson, P.J. B ristol. (1997) Targeting and translocation of proteins into the C a v a l i e r -S m i t h , T. (I986A). Cilia versus undulipodia. Bioscience hydrogcnosome of the protis! Trichomonas: similarities with 36, 293-294. mitochondrial proicin import. EMRO (European Molecular C a v a l i e r -S m i t h , T. (1986c). The kingdoms of organisms. Biology Organization ) Journal 16, 3484—3493. N ature 3 2 4 , 4 1 6 - 4 1 7 . B r i d g e , D ., Cunningham , C. W ., D eSalle, R & Buss, L. D. C a v a l i e r -S m i t h , T. (1987a). The origin of cells, a symbiosis VW ( 1995). Class-level relationships in the phylum Cnidaria: between genes, catalysts and membranes. Cold Spring Harbor molecular and morphological evidence. Molecular Biology and Evolution 12, 6 7 9 -6 8 9 . Symposia on Quantitative Biology 52, 805-824. C a v a l i e r -S m i t h , T . (1987 b). The origin of Fungi and pseudo­ B rugerolle, G . & J o y o n , L . (1975). Etude cvtologique ullraslruclurale des genres Proteromonas e t Karotomorpha (Z o o ­ fu n g i I n Evolutionary Biology o f the Fungi (ed. A. D. M .Rayncr, mastigophorea, Proteromonadida Grasse 1952). Protistologica C. M. Brasier and D. M. M oore), (Symp. Br. Mycol. Soc. 13), 11, 5 3 1 -5 4 6 . pp. 339-353. Cambridge University Press. C a v a l i e r -S m i t h , T . (1 9 8 7 cj. T h e o rigin o feukaryote and B r u n s , T . D ., W h i t e , T . J . & T a y l o r , J . W . 1 1 9 9 1 ). F u n g al molecular systematics. Annual Review o f 22, 5 2 5 -5 6 4 . archaebacterial cells. Annals o f the N ew York Academy of Sciences

B u i , E . T . N . , B r a d l e y , P. J. & Johnson, P .J . (1 9 9 6 ). A 503, 17-54. common evolutionary origin for mitochondria and hydro- C a v a l i e r -S m i t h , T. (1987tf). The simultaneous symbiotic genosomes. Proceedings o f the National Academy o f Sciences, USA origin of mitochondria, chloroplasts, and microbodies. Annals 93,9651-9656. of the .Sew fork Academy o f Sciences 503, 55 71.

B u l t , C . J ., W hite, O., O lsen, G. J ., Z hou, I... Kleischm ann, C a v a l i e r -S m i t h , T. (1989a). Systems of kingdoms. In M cC raw

R . D., Sutton, G. G., Blake, J . A., Fitzgerald, L. M., HUI Yearbook o f Science and Technology, pp. 175-179.

Clayton, R. A., Gocayne, J . D„ K erlavage, A. R., C a v a l i e r -S m i t h , T. (1989A). The kingdom Chromista. In The

D ougherty, B. A., T om b.J. F., Adams, M . D., R eich, C. I., Chromophyte Algae, Problems und Perspectives (ed . J . C . G re e n , B.

O verbeek, R , Kirkness, E. F., W einstock, K. G., M errick, S. C. I.eadbeater and W. C. Diver), pp. 381-407. Clarendon

J . M., Glodek, A., Scotr, J L., G eoghagen, N. S. M., Press, Oxford.

W e i d m a n , J . F., Fuhrm ann, J . I.., Nguyen, D., U tterback, C a v a l i e r -S m i t h , T . ( 1 9 9 0 ) . Symbiotic origin of peroxisomes. In

T . R., K elley, J. M., Peterson, J . D ., Sadow, P. W ., Endocytobiology IV (cd. P. Nardon, V. Gianinazzi-Pearson, A.

H ann, M . C., C otton, M . D., Roberts, K. M ., H urst, M. M. Grenier, L. Margulis and D. C. Smith), pp. 515-521.

A.. K aine, B. P., B orodovsky, M .. K lenk, H. P., F raser, C. P aris.

M ., Smith, H. O., W oese, C. R. & V e n t e r , J . C. 11996). C a v a l i e r -S m i t h , T. ( 1991 a). The evolution of cells. In Evolution Complete genome sequence of the melhanogenic archaeon, o f I.if e led. S. Osawa and T. Honjo), pp. 271-304. Springer- Methanococcus jannaschii. Science 2 7 3 ,1 05 8 -1 0 7 3 . Verlag, Tokyo. A revised six-kingdom system o f life 261

C a v a l i e r -S m i t h , T . ¡ 1 9 9 1 4 ) . Evolution of prokaryotic a n d fungi, and choanoflagcllates. Proceedings o f the Royal Society of eukaryotic cells. I n Foundations o f Medical Cell Biology ( e d . G. London B 261, 1 -6 .

E. Bittar), Vol. I, pp. 221—278. J.A .I. Press, Greenwich, CT, C avalier-Sm ith, T . & C h a o , E. E. (1996a;. Molecular USA. phylogeny of the free-living archezoan Trepomonas agilis a n d

C a v a l i e r -S m i t h , T . (19 9 1c ). Intron phylogeny: a new hy­ the nature <>l the first eukaryote. Journal o f Molecular Evolution p o th esis. Trends in 7, 145-148. 43,5 5 1 -5 6 2 .

C a v a l i e r -S m i t h , T . ( 1 9 9 1 d ) . Cell diversification in hetero­ C a v a l i e r -S m i t h , T . & C h a o , F.. F.. (1 9 9 6 4 ). 18S rR N A trophic flagellates. In The Biology o f Free-living Heterotrophic sequence of Heterosigma carterae (Raphidophyceae), and the Flagellates (ed. D .J. Patterson & J Larsen), pp. 113-131. phylogeny of heterokont algae (Ochrophyta).Phycologia 35, O x fo rd . 5 0 0 -5 1 0 .

C a v a l i e r -S m i t h , T. (1992 a). Origins of secondary metabolism. C a v a l i e r -S m i t h , T . & C h a o , E. E. (1997). Sarcomonad In Secondary Metabolites, their Function and Evolution (cd. D.J. ribosomal RNA sequences, rhizopod phylogeny, and the Chadwick and J. W helan), pp. 64-87. Wiley, Chichester. origin of euglyphid amoebae. Archiv für Protistenkunde 147,

C a v a l i e r -S m i t h , T. ¡19924). Bacteria and eukaryotes. N ature 2 2 7 -2 3 6 .

3 5 6 , 570. C a v a l i e r -S m i t h , T ., A l l s o p p , M. T. E. P Sí C h a o , E . E.

C a v a l i e r -S m i t h , T. (1993a). Kingdom Protozoa and its 18 (1994a). Chimeric conundra. are nucleomorphs and chro­ p h y la . Microbiological Reviews 57, 953-994. mists monophylclic or polvphyletic?Proceedings o f the National Cavalier-Smith, T. (19934). Evolution of the eukaryotic Academy of Sciences, USA 91, 11368-11372.

genome. In The Eukaryotic Genome, Organization and Regulation Cavalier-Sm ith, T ., A l l s o p p , M. T. E. P. & C h a o , E. E. (ed. P. M. A. Broda, S. G Oliver & P. F. G. Sims ), pp. ¡19944). Thraustochytrids arc chromists, not Fungi: I8S 333-385. Cam bridge University Press. rRNA signatures of Heterokonta. Philosophical Transactions of

C a v a l i e r -S m i t h , T. (1993c|. Percolozoa and the symbiotic the Royal Society of London B 33 9 , 13 9 -1 4 6 .

origin of the mclakaryote cell. In Endocytobiology V (e d . H . C a v a l i e r -S m i t h , T . C h a o , E. E . & A l l s o p p , M .T. E. P. Ishikawa, M. Ishida, and S. Satoi, pp. 399-406. Tubingen (1995). Ribosomal RNA evidence for chluroplasl loss within University Press. Heterokonta, pedinellid relationships and a revised classifi­

C a v a l i e r -S m i t h T. ( 1993(f)- The protozoan phylum Opalozoa. cation ofochristan algae. Archivfur Protiitenkunde 145,2 0 9 -2 2 0 .

Journal o f Eukaryotic Microbiology 40, 609-615. C avalier-Sm ith, T ., A l l s o p p , M. T. E. P., C h a o , K. K ,

C a v a l i e r -S m i t h , T. (1993rl. The origin, losses and gains of Boury-K snault, N . & V a c e l e t , J. (1996a). Sponge phy­ chloroplasts. In Origin o f Plastids: Symbiogenesis, Prochlorophytes logeny, animal monophyly and the origin of the nervous and the Origins o f Chloroplasts (ed. R. A. Lewin), pp. 291-348. system: 18S rRNA evidence. Canadian Journal o f foology 74, Chapman & Haii, New York. 2 0 3 1 -2 0 4 5 .

C a v a l i e r -S m i t h , T . ( 1 9 9 4 ) . Origin and relationships of Hapto­ C avalier-Sm ith, T ., C h a o , E . E ., T h o m p s o n , C . & H o u r i -

p h y ta . I n The Haptophyte Algae i.ed. J. C. Green and B. S. C. h a n e , S. (19964). Oikomonas , a distinctive zooflagellate related Leadbeateri, pp. 413—435. Clarendon Press, Oxford. lo chrysomonads. Archiv fü r Protistenkunde 146,2 7 3 -2 7 9 .

C a v a l i e r -S m i t h , T . ( 1 9 9 5 a ) . Membrane heredity, symbio­ C avalier-Sm ith, T ., C o u c h , J . A . , Thorsteixsen, K . E.,

genesis, and the multiple origins of algae. In Biodiversity and G i l s o n , P ., D e a n e , J . , H i l l , D . A . & M c F a d d e n , G . I. Evolution (ed. R. Arai, M. Kato and Y. Doi), pp. 75-114. (1996c). nuclear and nucleomorph 18S rRNA T o k y o . phylogeny.European Journal o f Phyeology 3 1 , 3 1 5 -3 2 8 .

C a v a l i e r -S m i t h , T . (19954). Cell cycles, diplokarvosis, and the C avalier-Sm ith, T ., A l l s o p p , M . T . F.. P ., H a u b e r , M . M .,

archezoan origin of sex. Archiv fu r Protistenkunde 145, 1 8 9 - 2 0 7 . R e x s i n g , S .A ., G o t h e , G ., C h a o , E. E ., C o u c h , J . A . &

C a v a l i e r -S m i t h , T. (1995c). Evolutionary comes M a i e r , U-G. ( 1996¿). Chromobiotc phvlogcnv: the enig­ o f ag e. Archiv fü r Prolistenkunde 145, 145-154. matic alga Reticulosphaera japonensis Grell is an aberrant

C a v a l i e r -S m i t h , T. ' 1995<í¡. Zooflagellate phylogeny and haptophyte, not a helerokont.European Journal o f Phyeology 31, classification. Cytology 37, 1010-1029 2 5 5 -2 6 3 .

C a v a l i e r -S m i t h , T. (1997a). Amocboflagellatcs and mitochon­ C h r i s t e n , R., R atto, A., B a r o i n , A ., P e r a s s o , R . G r e l l , K.

drial cristae in eukaryote evolution, megasyslcmatics of the G . & A d o u t t e , A. (1991). An analysis of the origin of new protozoan subkingdoms Eozoa and Neozoa. Archiv fü r mctazoans, using comparisons of partial sequences of the 28S Protistenkunde 147, 237-258. RNA, reveals an early emergence of triploblasts. EMBO

C a v a l i e r -S m i t h , T . 11 9 9 7 a ) . Sagenista and Bigyra, two phyla Journal 10, 499-503.

of heterotrophic heterokont chromists. Archiv Jür Protistenkunde C l a r k G . & R o g e r , A (1995). Direct evidence for secondary 148, 2 5 3 2 6 7 . loss of mitochondria in Entamoeba histolytica. Proceedings o f Ae

C a v a l i e r -S m i t h , T . ( 1 9 9 8 a ) . Ncomonada and the origin of National Academy of Sciences, USA 9 2 , 6 5 1 8 -6 5 2 1 .

animals and fungi. In Evolutionary Relationships Among Protozoa C l a r k , R. B. (1964). Dynamics in Melazoan Evolution. Clarendon :eds. G. H. Coombs, K. Vickcrman, M. A. Sleigh and A. Press, Oxford.

W arren), pp. 375—407 Chapman & Haii, London. C o h e n , B. I.., G a w t h r o p , A . B. & C avalier-Sm ith, T. i 1998).

C a v a l i e r -S m i t h , T. (19984). W hat are Fungi’ In: The M ycota M olecular phylogeny of brachiopods and phoronids based on 'cd. D. J. M cLaughlin), Vol. VII Springer-Verlag, in press. nudear-cncoded small subunit ribosomal RNA gene se­ C a v a l i e r -S m i t h , T . & A l l s o p p , M .T. E. P. (1996). Coral- q u e n ce s. Philosophical Transactions o f the Royal Society, B, in lochytrium, an enigmatic non-flagellate protozoan related lo press.

choanoflagcllates. European Journal of Protistology 32, 1 -9 . C o h n , F. (1875). Untersuchungen über Bactérien II. Beitrage zur C a v a l i e r -S m i t h , T . & C h a o , E. E. '1995). The opalozoa» Biologie der Pflanzen 1, Heft 3, 141-207

Apusomonas is related lo the common ancestor of animals, C o l e r i d g e , S. T. ¡1839) Aids to Reflection xv 64. 262 T. Cavalier-Smith

Conw ay M orris, S ., G e o r g e , J . D ., G i b s o n , R . & P l a t t , H . Fleischm an, R . D ., A d a m s , M . D ., W h i t e , O ., C l a y t o n , R . A .,

M. (1985). The Origins and Relationship ! o f Lower Invertebrates. K i r k n e s s , E. F ., K erlavage, A . R ., B u l t , C . J ., T o m b , J . F.,

Clarendon Press, Oxford. D ougherty, B. A ., M e r r i c k , J . M ., M c k e n n e y . K ., S u t i o n ,

C o p e l a n d , H . K. (1 9 5 6 ). The Classification of Lower Organisms. G ., F i t z h u g h , W ., F i e l d s , C ., G o c a y n e , J . D ., S u o t t , J .,

Pacific Books, Palo Allo. S h i r l e y , R ., L i u , I .. I., G l o d e k , A ., K e l l e y , J . M .,

C o r l i s s , J . O . (1971). Common sense and courtesy in nomen­ W e i d m a n , J . F ., P h i l l i p s , C . A., S p r i c g s , T ., H e d b l o m , E.,

clatura! taxonom y. Systematic 2 0 , 1 1 7 -1 2 2 . C o t t o n , M . D ., C tterback, T . R ., H a n n a , M . C ., N g u y e n ,

C o r l i s s , J. O. (1984). The kingdom Protista and its 45 phyla. D . T „ S a u d e k , D .M ., B r a n d o n , R . C ., F i n e , L . D .,

BioSystems 17, 8 7 - 1 2 6 . Fritchm an, J . 1.., F u h r m a n n , J . L ., G eoghagen, N. S. M .,

C o r l i s s , J. O. (1994). An interim utilitarian ('user friendly’) G n e h m , C . L ., M c d o n a l d , I,. A ., S m a l l , K . V ., F r a s e r , C .

hierarchical classification and characterization of the protists. M ., S m i t h , H . O . & V e n t e r , J C. (1995) Whole-gcnomc Acia Protozoologica 33, 1 - 5 1 . random sequencing and assembly of Haemophilus influenzae R d . C u m m i n g s , B. P . ( 1 9 1 6 ) . Studies on the Anoplura and Mallo­ Science 26 9 , 4 9 6 -5 ! 2. phaga, being a report upon a collection from the mammals Forterre, P., Benachenhou-Lahfa, N., Confalomeri, F., and birds in the society’s gardens. Part 1 with a preface. Duguet, M., Elie, C. & Labedan, B. (1993). The nature of ProceeSngs of the Zoological Society of London 1916 P a rt 1, the last universal ancestor and the root of the tree of life, still 2 5 3 - 2 9 5 . open questions BioSystems 28, 1 5 -2 3 . D avidson, L. A . ( 1 9 8 2 ) . Ultrastruclurc, behavior, and algal F r a s e r , C . M ., G o c a y n e , J . D ., W h i t e , O ., A d a m s , M . D ., flagellate affinities of the helioflagellatc Ciliophrys marina, and C l a y t o n , R . A ., Fleischm ann, R . D ., B u l t , C . J ., the classification of the helioflagellates (Protista, A dinopoda, K erlavage, A. R ., S u t t o n , G ., K e l l e y , J . M ., Fritchm an, Heliozoca), Journal o f Protozoology 29, 1 9 - 2 9 . J . L ., W e i d m a n , J . F ., S m a l l , K . V ., S a n d u s k y , M ., D e D u v e , C. (1982). Peroxisomes and related particles in F u h r m a n n , J . , N g u y e n , D ., U tterback, T . R ., S a u d e k , historical perspective.Annals o f the New York Academy o f Sciences D . M ., P h i l l i p s , C . A ., M e r r i c k , J . M ., T o m b , J - F ., 3 8 6 , 1 -t . D ougherty, R . A ., B ori-, K F., Hu, P-C., L u c i e r , T . S., D e s p o r t e s I, N a s h e d NN (1983) L’ltrastucturc of sporulation P e t e r s o n , S. N., Smith, H . O ., H u t c h i s o n I I I , C . A. & in Minchinia dentali (Arvy), an haplosporcan parasite of V e n t e r , J. C. (1995). The minimal gene complement of Dentalium entale (Scaphopoda, Mollusca) : taxonomic implica­ Mycoplasma genitahum. Science 27 0 , 397—f03. tions. Protistologica 19, 453—160. F u n c h , P. & K r i s t e n s e n , R. M . (1995). Cycliophora is a new D i M a r i a , P ., P a l i c , B., D ebrunner-V ossbrinck, B. A ., L a p p , phylum with affinities to F.ntoprocta and Ectoprocta.Nature J . & V ossbrinck, C. R. (1996). Characterization of the highly {¡sindon) 378,7 1 1 -7 1 4 . divergent U2 homolog in the microsporidian Vairimopha G a r e y , J R ., K r o t e c , M ., N e l s o n , D . R . & B r o o k s , J . necatrix. Nucleic Acids Research 2 4 , 5 1 5 - 5 2 2 . (1996a). M olecular analysis supports a tardigradc-arthropod D ourn, A. (1875). Ursprung der Wirbelthiere und das Princip des association. Invertebrate Biology 115,7 9 -8 8 . Functionwechsels . genealogischen skizzen. Engclmann, Leipzig. G a r e y , J . R ., N e a r , T . J . , N o n n e m a c h e r , M . R . & N a d l e r , S. D oolittle, W . F . & B r o w n , J. R. (1994). Tempo, mode, the A. (19965). Molecular evidence for Acanthocephala as a progenote, and the universal root. Proceedings oj the National subtaxon of Rotifera. Journal of Molecular Evolution 43, Academy o f Sciences, USA 91, 6 7 2 1 -6 7 2 8 . 2 8 7 -2 9 2 . E c h l i n , P. & M o r r i s , I. ( 1 9 6 5 ) . The relationship between blue- G e r m o t , A . , P h i l i p p e , H . , & L e G u y a d e r , H. (1996). Presence green algae and bacteria. Biological Reviews 4 0 , 1 4 3 - 187. of a mitochondrial-type 70-kDa heat shock protein in E d g e l l , D . R . & D oolittle, VV. F. (1997). Archaea and the origin(s) of DNA replication proteins. Cell 8 9 , 9 9 5 -9 9 8 Trichomonas vaginalis suggests a very early mitochondrial endosymbiosis in eukaryotes. Proceedings o f the National Academy E d l i n d , T . D ., L t, J ., V isvesvara, G . S ., V o d k i n , M . H ., o f Sciences, U SA 93, 14614-14617. M cLaughlin, G . I.. & K atiya, S. K . ( 1 9 9 6 ) . Phylogenetic analysis of beta tubulin sequences from amilochondrial G e r m o t , A ., P h i l i p p e , H . & L e G u y a d e r , H. (1997). Evidence p ro to z o a . Molecular Phylogeny and Evolution 5, 359-367. for loss of mitochondria in Microsporidia from a mito-

E d m a n , J . C ., K o v a c s , J . A ., M a s i .!r , H ., S a n t i , D . V . , E l w o o d , chondrial-type HSP70 in Nosema locustae. Molecular and Biochemical Parasitology 8 7 , 15 9 -1 6 8 . H . J . & S o g i n , M. I.. (1988). Ribosomal RNA sequence show s Pneumocystis cannii to be a member of the fungi. Nature G i b b s , S. P. (1981). The chloroplasts of some algal groups may 334, 519-22. have evolved from endosvinbiotic eukaryotic algae. Annals

E m b l e y , T . M., H irt, R. P. & W illiams, D .M . (1995). New York Academy o f Sciences 3 6 , 1 19 3 -2 0 8 . Biodiversity at the molecular level: the domains, kingdoms G i b b o n s R . Sz M u r r a y , R. G. E. (1978). Proposal concerning and phyla of life. In Biodiversity: Measurement and Estimation the higher taxa of bacteria. International Journal oj Systematic (ed. D. E. Hawksworth), pp. 21-33. Chapman & Haii, Bacteriology 28, 1-6. T.ondon. G i b b s , S. P. (1981). Chloroplasts of some groups may have

E n d e r l e i n G. (1925). Bakterien-Cyclogenie, B erlin. evolved from endosymbiotic eukaryotic algae. Annals o f the

F e d o r o v , O .V . , Pyatibratov, M . G ., K ostyukova, A . S., New York Academy o f Sciences 3 6 1 , 193 -2 0 7 .

O s i n a , N . K . & T a r a s o v , V. Yu (1994). Protofilamenl as a G ogarten.J. P ., K i b a k , H., DrrrRtuH, P. T a i z , I... B o w m a n , B.

structural clement of flagella of haloalkalophilic archac- J ., M a n o l s o n , M . F ., P o o l e , R . J . , D a t e , T ., O s h i m a , T .,

b a c tc ria . Canadian Journal of Microbiology 40, 4 5 -5 3 . K o n i s h i , J ., D e n d a . K & O s h i d a , M. (1989). Evolution of

Felsenstein, J. (1978). Cases in which parsimony or com­ the vacuolar H'-ATPasc, implications for the origin of patibility methods will be positively misleading. Systematic eukaryotes. Proceedings o f the National Academy of Sciences, USA Zoology 2 7 , 4 0 1 —4 1 0 . 68, 6661—6685. A revised six-kingdom system o f life 263

G oldfuss, G. A . ( 1 8 1 7 ) . L eber die Entwicklungshefen der Thiire. I s h i d a , K ., C a o , Y -, H a s e g a w a , M ., O k a d a , N . & H a r a . Y. Nürnberg, pp. 58. (1998). The origin of plastids, as inferred G o u l d , J . & K e e t o n (1 9 9 6 ). Biological Science , sixth ed. Norton, from phylogenetic comparisons of amino acid sequences of

N e w Y o rk . E F -T u . Journal oj Molecular Evolution 4 5 ( 6 ) in press.

G o l d a c r e , R .J . (1958). Surface films, their collapse on I w a b e , N’., K u m a , K ., H a s e g a w a , M ., O s a w a , S. & M i y a t a , T . compression, the shapes and sizes of cells and the origins oflifc. (1989). Evolutionary relationship of Archaebacteria, Eubac- I n Surface Phenomena in Chemistry and Biology (ed. J. F. Danielii, teria, and eukaryotes inferred from phylogenetic trees of K. G.A. Parkhurst and A C. Riddifin), pp. 278-298. duplicated genes. Proceedings o f the National Academy o f Sciences, Pergamon Press, New York. USA 86, 9355-9359.

G r a h a m , L E. (1993). Origin of Land Plants. Wiley, New York. J e f f r e y , C. ¡1971). Thallophytes and kingdom s-a critique.

G r e l l , K G., Heini, A . & Schuller, S. (1990). The Kew Bulletin 2 5 , 2 9 1 -9 ultrastructurc of Reticulosphaera socialis Grell {Hetero­ K a t a y a m a , T . H ., W a d a , H ., F u r u y a , H , S a t o h , N . & kontophyta). European Journal o f Protistology 26, 3 7 -5 4 . Y a m a m o t o (1995). Phylogenetic position of the dicyemid H a e c k e l , E (1 8 6 6 ). Generelle Morphologie der Organismen. R eim er, Mesozoa inferred from 185 r D N A sequences. Biological Bulletin

B erlin. 1 8 9 , 8 1 - 9 0 . H a f . c k e l , E. ¡1868). Natürliche Schöpfungsgeschichte. R e im e r, K e e l i n o , P. & D o o l i t t l e , W . F. ( 1996). a-Tubulins from early B erlin. diverging eukarvolic lineages : divergence and evolution of the H a n e l t , B ., van Schyndel, D ., A d e m a , C . M ., I . e w i s , I.. A . & tubulin family. Molecular Biology and Evolution 1 3 , 1 2 9 7 -1 3 0 5 . I.O K ER , E S. (1996). The phylogenetic position ofRhopalura K e e l i n g , P. & D o o l i t t l e , W. F. (1997). Evidence that ophiocomae () based on 18S ribosomal DNA eukaryotic triosephospbate isomerase is of alpha-prolco- sequence analysis. Molecular Biology and Evolution 13, 1187- bacterial origin. Proceedings o f the National Academy o f Sciences, 1191. USA 9 4 , 1 2 7 0 -1 2 7 5 . H aroraevE-s W . R . & D e a m e r , D. W. (1978). Origin and early K e e l i n g , P ., C h a r l e b o i s , R . L . & D o o l i t t l e , W. F. (1994). evolution of bilayer membranes. In Light Transducing Mem­ Archaebactcrial genomes: cubacterial form, eukaryotic con­ branes . Structure, Function and Evolution (cd. D. W. Deamer), pp. te n t. Current Opinion in Genetics and Development 4 , 8 1 6 -8 2 2 . 23-59. Academic Press, New York. K e n t , W. S. '1880-1882;. A Manual o f the Infusoria, Vols. 1 & 2. H ä u b e r , M . M , - M u l l e r , S . B ., S p e t h , V . & M a t e r , L '.-G . Bogue, I emdon. (1994). How to evolve a complex plastid? - A hypothesis. K i m u r a , M. (1983). The Neutral Theory o f Molecular Evolution. Botanica Acia 1 0 7 ,3 8 3 -3 8 6 Cambridge University Press. H aw ksw orth, D.E. (1995). Steps along the way to a Lake, J. A. (1988). Origin of the eukaryotic nucleus determined harmonised bionomcnclature. Taxon 44, 447—456. by rate invariant analysis of rRNA sequences. N ature 3 3 1 , H aw ksw orth, D . E ., K i r k , P . M ., S m i t h , B. C . & P e g l e r D. 18 4 -1 8 6 . N. ( 1995). Ainsworth and Bisby’s Dictionary o f the Fungi, 8 th ed. Lake, J. A. (1989). Origin of eukaryotic nucleus: eukaryotes CAB International, Wallingford. and cocvtcs are genotypically related. Canadian Journal of H e n n i g , W. (1966). Phylogenetic Systematics. University of Illinois Microbiology 3 5 , 1 0 9 -1 1 8 . Press, Urbana. L a m a r c k . J B. (1810). Philosophie fanlogique. Paris. H i b b e r d , D .J . ( 1 9 7 9 ) . The structure and phylogenetic signifi­ L a n g , B. F ., B u r g e r , G ., O ’K e l l y , C .J ., C e d e r g r e n , R ., cance of the flagellar transition region in the chlorophyll G o l d i n g , G . B ., L e m i e u x C . , S a n x o f f , D . , T u r m ï l , M & .--containing algae. BioSystems 11, 243- 261. G r a y , M. W. (1997). An ancestral mitochondrial DNA H i l l i s , D .M ., Huflsf.nbeck, J. P. & Cunningham, C.W . (1994). Application and accuracy of molecular phylogenies. resembling a eubacterial genome in miniature. Nature (London ) Science 2 6 4 ,6 7 1 -6 7 7 . 3 8 7 , 4 9 3 -1 9 7 . L a n k e s t E R , E. R. (1877). Notes on the embryology and H i n k l e , G . I . e i p e , D . D ., N e r a d , T . A. & S o g i n M . 1.. 1 1994). The unusually long small subunit ribosomal RNA of classification of the animal kingdom. Quarterly Journal of Phreatamoeba balamuthi. Nucleic Acids Research 22. 465—469. Microscopical Science, 1 7 , 3 9 5 -4 5 4 . L a n k e s t e r , E R . ( 1 8 8 0 ) . Degeneration. L o n d o n . Hori, H. T., Iron, T. & O s a w a , S. (1982). The phylogenetic structure of the metabacteria, Jenlralblattfü r Bakteriologie und L a n k e s t e r , F., R . (1 9 1 1 ). Z o o lo g y . I n Encyclopedia Britannica, Hygiene fC ] 3 , 18 -3 0 . 11th ed. V'ol. 28, pp. 1022-1039. Cam bridge University Press.

H o r n e r , D . S ., H i r t , R . P., K ilvinoton, S ., I . l o y d , D . & Lee, J.J. (1990). Phylum Granuloreticulosa (Foraminifera). In

E m b l e y , T. M. (1996). Molecular data suggest an early Handbook oj Protoclisla (cd. L. Margulis, J. O. Corliss, M acquisition of the mitochondrion endosymbiont. Proceedings of Mclkonian and D .J. Chapman), pp. 524-548. Jones & the Royal Society o f London B 2 6 3 , 10 5 3 -1 0 5 9 . Bartlett, Boston. H u x l e y , J. S. (1957). The three types of evolutionary process. I.E E D A L E , C. F. 1,1974). How many arc the kingdoms of N ature, 180, 454-455. organisms? Taxon 3 2 , 2 6 1 -2 7 0 .

H u x l e y , J S. (1959). Clados a n d grades. In Function and L e e u w e n h o e k , A. v a n (1 6 7 5 ). Philosophical Transactions o f the Taxonomic Importance icd A.J. Cain), pp. 21-22. The Sys- Royal Society o f London.

tematics Association, London. L e e u w e n h o e k , A . v a n (1 6 8 3 ). Philosophical Transactions of the

H y m a n , L. H. (1940). The Invertebrates, V o l 1. M c G ra w -H ill, Royal Society o f London.

N e w Y ork. L e i p e , D . D ., W a i n r i g h t , P . O ., G u n d e r s o n , J . H ., P o r t e r ,

H y m a n , L. H. (1951). The Invertebrates, Vol. 3. McGraw-Hill, D ., P a t t e r s o n , D . J ., V a l o i s e , F , H i m m e r i c h . S. & S o g i n , N e w Y o rk . M L . 1 1994). The stramenopila from a molecular per­

H ym an, I.. H. (1959). The Invertebrates, V o l. 5 . McGraw-Hill, spective, 165-like rRNA sequences from ¡Mbyrinthuloides minuta

N e w Y o rk . a n d . Phycologia 3 3 , 3 6 9 -3 7 7 . 2 6 4 T. Cavalier-Smith

L e m e r y , N . ( 1 6 7 5 ) . Cours de Chymie contenant la maniere de faire les M ereschkowsKY, C. (1910). Theorie der zwei Plasmaarten als operations pui soni en usage dans la medecine , par une methode facile Grundlage der Symhiogenesis, einer neuen Lehre von der avec des raisonnements chaque operation, pour Pinstruction de ceux qui Entstehung der Organismen. Biologisches Centralblatl, 30, veulent s'appliquer a celle science. Lemery, Paris. 278-303, 321-347, 353-367.

L e v i n e , N. D. (1970). Taxonomy of the Sporozoa. Journal o f M ikrjukOv, K . A. & M y l n i k o v , A. P. ( 1995). Fine structure of Parasitology 56 (sec. O,), 2 0 8 -2 0 9 . an unusual rhizopod, Penardia cometa, containing extrusomes L ichtw ardt, R. W. ! 1988). The Trichomycetes, Fungal Associates and kinclosomcs. European Journal o f Protistology 31, 90-96.

of Arthropods. Springer-Verlag, New York. M ikryukov, K . A . & M y l n i k o v , A P. (1996a). Protist M u lti­

L í J . , K a t i y a r , S. K ., H a m e l i n , A ., V i s v e s v a r a , G . S. & cilia marina Cienk. : flagellate or heliozoan’ Transactions o f the K d u n d , T. D. (1996). Tubulin genes from AIDS-associated Russian Academy o f Sciences 34 6 , 136-139 (in Russian)

microsporidia and implications tor phylogeny and benzi- M ikryukov, K . A & M y l n i k o v , A. P. (19966). New in­ midazole sensitvity. Molecular and Biochemical Parasitology 78, formation of structure and life cycle of Athalamida Amoebae 2 8 9 -2 9 5 . (Protista, Athalam ida). .Joo/ogico/ Journal 75, 1283-1293.

Lom, J. & De Pltytorac, R. (1965). Studies on the mvxo- M i n e l l i , A. (1995). Biological Systematics: The State o f the Art. sporidian ultraslrueturc and polar capsule development. Chapman & Haii, London.

Prolislologica 1, 5 3 -6 5 . M ö h n , E. (1984). System und Phylogenie der Lebewesen. B a n d 1.

L o r e n z e n , S. ( 1 9 8 5 ) . Phylogenetic aspects o f pscudocoelomate Schwertzcrbart'sche Verlag, Stuttgart.

evolution. In The Origins and Relationships o f Lower Invertebrates M oore-Landecker, E. (1996). Fundamentals o f the Fungi, 4 th ed. {cd. S. Conway Morris, J. D. George, R. Gibson and H. M. Prenlice-Hall, Englewood Cliffs. NJ

Platt), pp. 210-233. Clarendon Press, Oxford. M o r i n , L . & M i g n o t , J-P. (1996). Are Archamoebae true

M a c k e y , L. Y ., W innepenninckx , B., De W ächter, R., Archezoa? The phylogenetic position ofPelomyxa sp. as

B a c k e l j a u , T ., K m s c h e r m a n n , P. & G a r e y . J . R. (1996). 18S inferred from large subunit ribosomal RNA sequencing. rRNA suggests that (entoprocta are protostomes, unrelated to European Journal of Protistology in press.

Kctoprocla. Journal o f Molecular Evolution 42, 552-559. M ü l l e r , M . (1992). Energy metabolism ofancestral eukaryotes,

M a n t ó n , 1. (1965). Some phyletic implications of flagellar a hypothesis based on the biochemistry of amitoehondriate structure in plants. Advances in Botanical Research 2, 1-34. parasitic protists. BioSystems 28, 3 3 -4 0 .

M a n t ó n , S. M. (1977). The Arthropoda: Habits, Functional N a g a h a m a , T ., S a t o , H ., S h i m a z u , M . & S u g i y a m a , J. (1995). Morphology and Evolution. Oxford, Clarendon Press. Phylogenetic divergence of the entomophthoralean fungi:

M a r c h a n t , H . J . & T h o m s e n , H . ( 1 9 9 4 ) . Haptophytesi n p o la r evidence from nuclear 18S ribosomal RNA sequences. waters. In The Biology o f the Haptophytes (ed. J. C. Green and Mycologies 87,2 0 3 -2 0 9 .

B. Leadbeater), pp. 209-228. Clarendon Press, Oxford. N e c k e r , N .J . d e (1 7 8 3 ). Traite’ sur la mycitologie ou discours

M a r g u l i s , I,. (1970). Origin o f Eukaryotic Cells. Yale University historique sur les champignons en general, dans lequel on demontre leur Press, New Haven. veritable origine et leur génération ; d1 ou dependent les effets pernicieux

M a r g u l i s , L. (1981). Symbiosis in Cell Evolution. Freeman, San et funestes de ceux que l'on mange avec les moyens de les éviter. F ran c isco . M atthias Fontaine, Mannheim.

M a r g u l i s , L . & S c h w a r t z , K. V. (1988). Five Kingdoms, 2nd N e l s o n , G . & P l a t n i c k , N. (1981). Systematics and Biogeography : ed. Freeman, New York. and Vicariance. Columbia University Press, New

M a r v i n -S i k k e m a , F. D ., K r a a k , M . N ., V e e n h u i s , M ., G o t t - Y ork.

s c h a l , J . C . & P r i n s , R. A. (1993). The hvdrogenosomal N i e l s e n , C. (1995). Animal Evolution: Interrelationships o f the Living enzyme hvdrogenase from the anaerobic fungus Meocallimastix Phyla. Oxford University Press.

sp. 1.2 is recognised by antibodies, directed against the C- O l s e n , G . J . & W o e s e , C. R. (1993). Ribosomal RNA, a key to terminal microbodv protein targeting signal SKI.. European phylogeny.FASEB Journal 7, 113 -1 2 3 .

Journal oj Cell Biology 61, 86-91. O l s e n , G . J . & W o e s e , C . R . & O v e r b e e k , R. (1994). The

M a t t o x , K . R . & S t e w a r t , K . D . ( 1 9 8 4 ) . Classification o f th e winds of (evolutionary) change: breathing new life into green algae: a concept based on comparative cytology. In microbiology. Journal o f Bacteriology 176,1-6 .

Sytematics of the Green Algae (ed. D. E. G. Irvine and D. M. O u z o n i s , C . A . 8t K y r p i d e s , N .K . (1996). Hypothesis: the John), pp. 29-72. Academic Press, London. emergence of major cellular processes in evolution. FEBS

M a y n a r d S m i t h , J . & S z a t h m a r y , E. (1995). T he M ajor Letters 3 9 9 , 119 -1 2 3 .

Transitions in Evolution. Freeman, Oxford. O w e n , R. (1858). Palaeontology. Encyclopedia Britannica (8th

M a y r , E. (1990). A natural system of organisms. Mature 348, edn) (ed. T. S. Traill), Vol 17, 91 176. Edinburgh.

4 9 1 . P a g e , F. C. (1987). The classification of 'naked' amoebae

M a y r , E . & A s h l o c k , P .D . ( 1 9 9 1 ) . Principles o f Systematic (Phylum Rhizopoda). Archiv fü r Protistenkunde 133,199 -2 1 7 .

Zoology , 2nd edition. M cGraw Hill, New York. P a n t i n , C. F. A. (I960). Diploblaslic animals. Proceedings o f the

M c F a d d e n , G . L , W a l l e r , R . , R e i t h , M E. & I. a n g - Linnaean Society o f London 171,1-14.

U n n a s c h , N. (1997). Plastids in apicomplexati parasites. In Patterson, D. J. (1985). The fine structure of ranarum The Origins of Algae and their Plastids i ed. D. Bhattacharya). (family ). Opalinid phylogeny and classification. Springer-Verlag, Berlin. In press. Prolislologica 21, 4 1 3 -4 2 8 .

M e l k o n i a n , M ., M a r i n , B. & S u r e r , B. ( 1 9 9 5 ) . P h y lo g e n ya n d Paiterson, D .J. (1986). The aclinophryid heliozoa (Sarcodina, evolution of algae. In Biodiversity and Evolution (cd. R. Arai, M. Adinopoda) as chromophytcs. In Chrysophyles: Aspects and Kalo and Y. Doi), pp. 153-176. The National Science Problems (cd R. A. Andersen and J. Kristiansen), pp. 49-67. Museum Foundation, Tokyo. Cambridge University Press. A revised six-kingdom system o f life 265

Patterson, D .J. (1988). The evolution of Protozoa. Memoirs ution of . In Biodiversity and Evolution (cd. R. Arai, M. Institute Oswaldii Cruz, Rio de Janiero. Suppl. 1. Vol. 83, Kato and Y. Doi), pp. 267-290. National Science Museum 5 8 0 -6 0 0 . Foundation, Tokyo.

Patterson, D .J . ( 1 9 8 9 ) . Stramenopiles, chromophytes from a ScHNEPF, E. & S c h w e i k e r t , M. (1997). Pirsonia, phagotrophic protistan perspective. In The Chromophyte Algae, Problems and nanoilagcllalcs . feeding on marine diatoms; Perspectives (ed. J. C. Green, B. S. C. Leadbeater and VV. I„ host recognition, attachment and fine structure. Archiv fu r

Diver), pp. 3 5 7 - 3 7 9 . Clarendon Press, Oxford. Protistenkunde 147,3 6 1 -3 7 1 .

Patterson, D .J. (1994) Protozoa, evolution and systematics. S c h w e i k e r t , M . & S c h n e p F , E. (1997). I.ight and electron I n Progress in Protozoology (ed. K. Hausmann and N. microscopical observations on Pirsonia punctigerae sp. nov., a Huismanii), pp. 1-14. Fischer, Stuttgart. nanoflagellate feeding on the marine centric Thalas­

Patterson, D . J . & F e n c h e l , T. (1985). Insights into the siosira punctigera. European Journal o f Protistology 3 3 , 168—177. evolution of heliozoa (Protozoa, Sarcodina) as provided by S c h o p f , J . W . (1993). of the early Archean Apex ullraslruclural studies on a new species of flagellate from the chert: new evidence of the antiquity of life. Science 2 6 0 , g e n u s Pteridomonas. Biological Journal o j the Linnean Society 34, 6 4 0 -6 4 6 . S c h o p f , J . W . & K l e i n , C. ( 1 9 9 2 ) . The Prolerozoic Biosphere: a 3 8 1 - 4 0 3 . Multidisciplinary Study. Cambridge University Press. P a w l o w s k i , J . , B o l i v a r , I., G u i a r o -M o f f a , J . & G o u y , M . (1994). Phylogenetic position of Foraminifera inferred from S e d g e w i c k , A . ( 1 8 9 8 ) . A Student's Text-book o f Zoology, v o l . 1. I.SU rRNA gene sequences. Molecular Biology and Evolution 11, Sonnenschein, London. S e n s e n , C . W . , K l e n k , H . P . , S i n g h , R . K ., A l l a r d , C . , C h a n , 9 2 9 -9 3 8 . C. C. Y., Liu, Q, Y., P e n n y , S. L ., Y o u n g , F ., S c h e n k , M. P a w l o w s k i , J . , B o l i v a r , I ., F a h r n t , J . F ., C a v a l i e r -S m i t h , T . E ., G a a s t e r l a n d , T ., D o o l i t t l e , W . F . , R a g a n , M . A . & & G o u y , M. (1996). Early divergence of Foraminifera C h a r l e b o i s , R . L . ( 1 9 9 6 ) . Organizational characteristics and suggested by SSU rRNA gene sequences. Molecular Biology and information content of an archaea! genome: 1 5 6 k b o f Evolution 13, 4 4 5 -4 5 0 . sequence from Sulfolobus solfataricus P 2 . Molecular Micro­ Perty, J. A. M. (1852). ju t Kenntnis der Kleinster Lebensformen. biology 22,175-191. B ern. S h a r o v , A . G . ( 1 9 6 6 ) . Basic Arthropodan Stock with Special Reference P i c k e n , L. R. (1960). The Organization of Cells and Other to Insects. Pergam on, Oxford. Organisms. Oxford University Press. S i d d a l l , M . E ., S t o k e s , N . A. & B u r r e s o n , E.M (1995). P i e r s o n , B. K. & Olson, J . M . ( 1 9 8 9 ) . E v o lu tio n o f p h o to ­ Molecular phylogenetic evidence that the phylum haplo­ synthesis in anoxygenic photosvnthelic procaryotes. In sporidia has an alveolate ancestry. Molecular Biology Sc Microbial Mats, Physiological Ecology of Benthic Microbial Evolution 12, 5 7 3 -5 8 1 . Communities (ed. Y. Cohen and E Rosenberg), pp. 402—427. S i d d a l l , M . E ., M a r t i n , D . S. B r i d g e , D . , D e s s e r , S. S . & American Society for Microbiology, W ashington. C o n e , D . K . ( 1 9 9 5 ) . The demise of a phylum of protists: R a j k o v , I. B. ¡1982). The Protozoan Nucleus, Morphology and Phylogeny of Myxozoa and other parasitic cnidaria. Journalo f Evolution. Springer. Wien. Parasitology 8 1 ,9 6 1 - 9 6 7 . Raohu-Kumar, S. i 1987). Occurrence of the thraustochytrid, S i l b e r m a n , J . D ., S o g i n , M . L . , L e i p e , D D . Sc C l a r k , C . G Corallochytrium Itmacisporum gen. ct sp. nov. in the coral reef 11996). H um an parasite finds taxonom ic hom e. N ature 3 8 0 , lagoons of the l.akshadweep Islands in the Arabian Sea. 398. Botanica Marina 3 0 , 83 89. S i m p s o n , A G . B . , v a n d e n H o f f . J . , B e r n a r d , C ., B u r t o n , H . R a g a n , M . A . & G u t e l l , R. R. (1995). Are red algae plants? R . & P a t t e r s o n , D . J . H997). The ullrastructurc and Botanical Journal o f the Linnean Society 118, 81-105. systematic position of the cuglcnozoon Postgaardi mariagerensis. R a v e n , P. H. (1970). A multiple origin for plastids and F e n c h e l et al. Archiv fü r Protistenkunde 147,2 1 3 -2 2 5 . mitochondria. Science 169, 6 4 1 -6 4 6 . S i n g e r , C . ( 1 9 3 1 ) . A Short History o f Biology. Clarendon Press, Reisinger, E. (1938). Pogonofora. Fortschritte Zoologische 3 , for O x fo rd . 1937. S m o t h e r s , J . F ., V o s D o h l e n , C . D . , S m i t h , L. H. Jr. & R i e g e r , R. (1985). The phylogenetic status of the acoelomate S p a l l , R. O. (1994) Molecular evidence that the myxozoan organization within the Bilateria; a historical perspective. In protists are metazoans. Science 265, 1719—1721. The Origins and Relationships of Lower Invertebrates (ed S. S o l t y s , B .J . & G u p t a , R . S. (1994). Presence and cellular Conway Morris, J. D. George, R. Gibson and H. M. Platt), distribution of a 60 kDa protein related to mitochondrial pp. 101-122. Clarendon Press, Oxford. h sp 6 0 in Giardia ¡amblia. Journal of Parasitology 8 0 , 5 8 0 -5 9 0 . R o g e r , A. J. ( 1996). Studies on the Phylogeny and Gene Structure of S n o d g r a s s , R. E. (1938). Evolution of the Annelida, Onycho­ Early Branching Eukaryotes. Ph.D. thesis, Dalhousie University. phora and Arthropoda. Smithsonian Miscellaneous Collection 97, C a n a d a . 1-1 5 9 .

R o g e r , A . J ., G raham C lark, C , D o o u i t l e , W. F. (1996) A S tackebrandt , E. & W o e s e , C . R . ( 1 9 8 1 ) . The evolution of possible mitochondrial gene in the early-branching amito- prokarvotes. In Molecular and Cellular Aspects of Microbial chondriate prolist Trichomonas vaginalis. Proceedings of the Evolution (cd. M .J. Carlile, J. F. Collins and B. E. B.

National Academy of Sciences, USA 93, 14618-14622. Moselev), pp. 1—3 1 . Cambridge University Press.

Rusk, S., Spiegel, F. & Lee, S. (1994). Phylogenetic relation­ S t a n i e r , R .Y . ( 1 9 6 1 ) . La place des bactéries dans le monde ships of slime moulds inferred from ribosomal DNA. Abstracts v iv a n t. Annales d Institut Pasteur 101, 2 9 7 -3 1 2 .

Fifth International Mycological Congress, Vancouver, 1 8 4 . S t a n i e r , R. ! 1970). Some aspects of the biology of cells and their S a t OH, N. (1995). Towards a molecular understanding of possible evolutionary significance. Symposium for the Society oj

developmental mechanisms underlying the origin and evol­ Generna! Microbiology 2 0 . 1 - 3 8 . 266 T. Cavalier-Smith

S t a n i e f , R . Y., D oudoroff, M . & A d e l b e r g , F,. A. ; 1963). Von A rx, J. A. 1 1967). Pilzkunde. Cramer, Lehre

The Microbial World, 2nd cd. Picntkc-Hall, Englewood Cliffs, SVa n o , X . & L u t k e n h a i i s , J . 11996). Fis Z ring: the cubaclcrial

KJ division apparatus conserved in archaebacteria. Molecular S t a n t e r , R. Y. & V a n N i e i , C. B. ' 1 9 6 2 ) . T h e concept of a Microbiology 2 1 , 3 1 3 -3 1 9 .

bacterium . Archiv ¡ür Mikrobiologie 4 2 , 1 7 - 3 5 . W ainw right, P. O ., H i n k l e , G ., S o g i n , M . L. & S t i c k e l , S. K . S t i l l e r , J , W . & H a l l , B. D . ( 1 9 9 7 ) . T h e origin of red algae: (1993). M onophyletic origins of the M etazoa. an evolutionary im plications for plastid evolution. Proceedings oj the .Xahoctal link with Fungi. Science 26 0 , 3 4 0 -3 4 2 . Academy o f Sciences, U SA 94, 4520-4525. SV e i l l , R (1938). L'interprétation des Cnidosporidies et la S t u e s s y , T. F. ¡1990}- Plant Taxonomy. Columbia University valeur laxonomique de leur cnidome. Leur cycle comparé à la Press, New York. phase laivairt des NarcomédusesCuninidcs. Travaux du Slalom S i u n k a r d , H. W. (1954). The life history and systematic foologique Wimereux 13, 7 2 7 -7 4 4 . relations of the Mesozoa. Quarterly Review o f Biology 29, W h a t l e y J . M ., J o h n P . & W h a t l e y F. R. (1979). From 230-244. extracellular to intracellular: the establishment of mito­ S i u n k a r d , H . \V. ¡1972). Clarification of taxonomy in Mesozoa. chondria and chloroplasts. Proceedings o f the Royal Society of Systematic f oology 21, 210-214. London Series B 2 0 4 , 165 -1 8 7 . Suoiyam a J. & N i s h i d a , H. (1995). The higher Fungi: their evolutioarv relationships and implications for fungal system­ W hittaker, R. H. (1969). New concepts of kingdoms of atics. In Biodiversity and Evolution (ed. R Arai, M.Kalo and Y. organisms. Science 1 6 3 , 15 0 -1 6 0 . Doi), pp. 177-195. National Science Museum Foundation, W innepenninckx , B . , B a c .k e l j a u , T ,, M a c k e y , I . . , B r o o k s , J , Tokyo. M ., D e W a c h t e r , R . D ., K u m a r , S. & G a r e y , J. R. (1995a).

S u o i y a m a , J ., N a g a h a m a , T . & N i s h i d a , H. (1996) Fungal 18S rRNA data indicate that Aschelminthes are polyphylctic diversity and phylogeny with emphasis on 18S ribosomal in origin and consist of at least three distinct clades. Molecular DNA sequence divergence. In Microbial Diversity in Time and Biology and Evolution 12, 113 2 -1 1 3 7 . Space (cd. R. Cowell, U. Simidu and K. Ohtvada), pp. VVlNNEPENNINCKX, B ., BaCKELJAU, T . & D e WÄCHTER, R . D . 150-160. Plenum, New York. (19953). Phylogeny of protostome worms derived from 18S

S w o f f o r d , D . I.,, O l s e n , C . J ., W addell, & H illis, D. (1996). rRNA sequences. Molecular Biology and Evolution 12, 6 4 1 -6 4 9 .

Phylogenetic inference. In Molecular Systematics, 2nd cd. (ed. W o e s e , C . R . 11979). A proposal concerning the origin of life on D. M. Hillis and C. Moritz), pp. 407-51. Sinauer, Sunder­ the planet earth. Journal of Molecular Evolution 12, 9 5 -1 0 0 .

lan d , M A . W o e s e , C. R. (1983). The primary lines of descent and the T aylor, F. J R. ¡1974). Implications and extensions of the universal ancestor. In Evolution from Molecules to Men (e d . D. serial endosymbiosis theory of the origin of eukaryotes. Taxon S. Bendall), pp 209-233. Cambridge University Press. 23, 229-258. W o e s e , C. R. (1987). Bacterial evolution. Microbiological Reviews T a y l o r , F.J. R. ¡1978). Problems in the development of an 5 1 , 2 2 1 -2 7 1 . explicit hypothetical phylogeny of the lower eukaryotes. W o e s e , C. R. (1994). There must be a prokaryote somewhere: BioSystems 10, 67-89. microbiology's search for itself. Microbiological Reviews 5 8 , 1 -9 . T h o m s e n , H . A. & L a r s e n , J. (1993). The ultrastruciure of W o e s e , C. R. & Fox, G E. (1977a). Phylogenetic structure of Commation gen. nov (stramenopiles incertae sedis), a genus of the prokaryotic domain : the prim ary kingdoms.Proceedings o f heterotrophic naiioplanktonic flagellates from Antarctic the Rational Academy o f Sciences, USA 74,5 0 8 8 -5 0 9 0 . w aters. European Journal o f Protistology 29, 462-477. W o e s e , C. R. & Fox, G. F.. (19773). The concept of cellular T o n g , S. M . ( 1 9 9 5 ) Developayella elegans, nov. gen., nov. spec., a evolution. Journal o f Molecular Evolution 10, 1-6 . new type ol heterotrophic flagellate from m arine plankton. W o e s e , C . R ., K a n d l e r , O . & W h e e l i s , M. L. (1990). Towards European Journal oj Protistology 31, 24-31. a natural system of organisms, proposal for the domains T u r n e r , S. ¡ 1 9 9 8 ) . Molecular systematics o f oxygenic photo- synthctic bacteria. In The Origin o f Algae and their Plastids (ed. Archaea, Bacteria, and Eucarva. Proceedings of the .Valional D. Bhallacharyaj. Springer-Verlag, Berlin, in press. Academy of Sciences, USA 8 7 ,4 5 7 6 -4 5 7 9 . W o e s e , C . R . S tackebrandt , F.. & L u d w i g , W. (1985). W hat U e s h i m a , R i , 1 9 9 5 ) . Evolution and phylogeny of the Metazoa with special reference to recent advances in molecular arc mvcoplasmas: the relationship of tempo and mode in phylogeny. InBiodiversity and Evolution (cd. R. Arai, M. Kalo bacterial evolution. Proceedings of the Xational Academy of and Y. Doi), pp. 197-228. National Science Museum Sciences, U SA 2 1 , 3 0 5 -3 1 6 .

Foundation, Tokyo. W r a y , C . G ., L a n g e r , M R .. D e s a l l e , R ., L e e , J . J . & I. i p p s ,

V on Sif.bold, C. T. ¡1845). Lehrbuch der Vergleichende Anatomie der J. H. (1995). Origin of the foraminifera. Proceedings o f the Wirbellosen Thiere Heft 1. pp. 10-11 Viel + Comp., Berlin. .\ational Academy i f Science, USA 9 2 , 14 1 -1 4 5 .