PhD and postdoctoral years
Between 1963 and 1966, Tom worked on his PhD in the laboratory of Sir John Randall FRS head of the Biophysics Department, University of London King’s College. Randall’s group studied eukaryotic cilia/flagella using as a model system the unicellular green alga Chlamydomonas reinhardtii. Tom’s thesis (1967) focused on organelle development in the green alga. He carried out detailed electron microscopic investigation of the behaviour of chloroplasts, mitochondria, centrioles and cilia throughout the sexual life cycle of Chlamydomonas. The introduction of glutaraldehyde fixation by David Sabatini in 1963 enabled higher-quality ultrastructural studies [1]. Tom’s main interest was to illuminate the problems of organelle heredity and non-Mendelian inheritance through a better characterisation of organelle structure across the entire life cycle, including mitosis, meiosis and the following gamete and organelle fusion.
Tom’s discoveries from this period included the continuity and fusion of the chloroplasts of the two gametes (one per gamete) into one zygotic chloroplast, thus providing a cytological basis for non-Mendelian heredity in Chlamydomonas [2]. His ultrastructural observations also suggested that centrioles have no genetic continuity throughout the life cycle and can be assembled de novo at the time of meiosis [3]. Further observations included a slender copulation tube that forms during gamete fusion as an extension of the plasma membrane [4] and the description of zygospore formation [5]. Tom was sole author on these papers (published only during the 70s). He only thanked Sir John Randall for encouragement and support, indicating how differently science worked in those days.
In 1966 Tom set sail with his first wife Gillian Glaysher and their two small children (Jane, 1964 and Neal, 1966) in a wallowing Italian ship, the Aurelia, from Southhampton to New York. In New York, Tom spent two intellectually stimulating years at Rockefeller University in George Palade’s Cell Biology Department working as a Guest Investigator on a Damon Runyon Memorial Fund Postdoctoral Fellowship in the laboratory of David Luck. He was especially intrigued by the cell biological and evolutionary significance of mitochondrial DNA and wanted to gain biochemical experience working with cell organelles to complement his ultrastructural studies. After some exploratory experiments with the filamentous fungus Neurospora, Tom returned to working on Chlamydomonas trying to isolate temperature sensitive motility mutants. The project turned out to be more difficult than expected and despite isolating nearly 100 temperature sensitive swimming mutants he did not have time to study any of them in detail. Although no significant research outputs emerged during these years, the time at Rockefeller was important to develop further experimental skills and a more in-depth understanding of cell biology. His colleagues at Rockefeller included Günter Blobel, who together with David Sabatini started to develop the signal hypothesis on intracellular ribosome targeting around this time [6].
From Kings College through Vancouver to Oxford
In 1969 Tom got a job as Lecturer of Biophysics at Kings College London and returned with his family to the UK on the Aurelia. He took on significant teaching commitments, including the running of the undergraduate Cell and Molecular Biology degree for the following 20 years.
After being awarded tenure in 1972, he continued to give most effort to teaching and administration. His research increasingly moved away from characterising Chamydomonas mutants to thinking about eukaryote cell and chromosomal structure and evolution. Later he learn how to characterise and sequence DNA in order to test some of his phylogenetic ideas more thoroughly.
Tom was promoted to Reader in 1982 and continued his research at Kings on genome size evolution and eukaryote cell evolution. He had separated from his wife in 1984 and spent his summer vacation in the Pacific North West of North America. He visited Max Taylor a marine biologist specialising in dinoflagellates of the University of British Columbia (UBC) in Vancouver, Canada. He really liked the environment and started to wonder if one day he may be able to move there. In 1985-1986 he spent a sabbatical year in Australia to first work with sponge expert Clive Wilkinson at the Australian Institute for Marine Sciences near Townsville, Queensland and then with Des Clark-Walker at the Australian National University in Canberra. Tom collected sponge and cnidarian samples from the Great Barrier Reef and isolated their DNA. During his sabbatical, he also met the German retired protozoologist Karl Grell, the rediscoverer of Trichoplax, the simplest animal.
The following year, he applied and received a position of Full Professor of Botany at UBC. The final straw that stimulated his move to Canada was Mrs Thatcher’s threat in late 1987 to put the UK universities more under her thumb. After his move, he became a Fellow of the Canadian Institute for Advanced Research (1989–1998). It was in Vancouver that Tom recruited in 1989 as graduate research assistant Ema E-Yung Chao, his second wife and collaborator. They got married in 1991 and had a daughter Rose in 1993. Ema was important for the smooth running of Tom’s lab for about 25 years and after its closure helped him with computer work until their last joint paper published in 2020 [7].
Not long before Tom left the UK, he had visited the NERC Culture Collection of Algae and Protozoa at Windermere and observed Petri dishes of single protozoan cultures such as Apusomonas “with hordes of identical cells wiggling across the dish”. This stimulated Tom “to want to sequence the 18S rDNA of as many protists whose taxonomy was uncertain and evolutionary relationships were obscure by using such uniform cultures.”
UBC gave Tom generous startup funds that enabled him to invest in DNA sequencing and microscopy. His sequencing programme really took off after they switched from manual radiolabelled sequencing to automated sequencing with fluorescent labelling. The sequencing started to result in publications from 1993 onward (including one on Apusomonas [8]). In 1993 he published a major taxonomic synthesis for protists. This phylogeny explicitly emphasised ultrastructure and included Tom’s first large-scale rDNA tree for 150 eukaryotes [9]. During the UBC years, Tom also wrote on many theoretical topics, including the evolution of peroxisomes (first considered symbiotic, later autogenous), a critical evaluation of the fossil evidence in cell evolution [10], and intron phylogeny [11].
In 1996 Ema and Tom moved to South Africa for a one-year sabbatical. His colleagues in South Africa recall happy memories of Tom “in his bush hat, shorts and hiking boots with his binoculars round his neck”.
Tom and Ema stayed at UBC until 1999 when Tom took up a NERC Professorial Fellowship at the Department of Zoology at the University of Oxford. The move back to the UK was party motivated by considerations about their daughter’s schooling, a preference for the old world, the prestige of Oxford and an optimistic picture of Britain becoming more integrated into the EU. About half of Tom’s total research publications came after the move to Oxford, reflecting the environment, the maturation of Tom’s ideas and a freedom from teaching as a NERC fellow. He continued to integrate morphology, sequence evolution, and large scale synthesis.
In 2016, Ema and Tom moved to Cornwall where they kept working together until Tom’s death in 2021. Tom remained a naturalist until his death. In Cornwall, they bought a 10-acre plot where they used to sit, watch and listen to the seasonal residential and migratory birds. Together they had planted over 200 native trees and shrubs to start their own little patch of nature conservation. Tom’s longest and — according to him — maybe most important papers were published in 2020 [7] and 2021 [12] and “these may have to substitute for a summarising book if my cancer does not allow me time and energy to write it”.
WORKS
Research activities and achievements
The research career of Thomas Cavalier-Smith spanned over 50 years, from the early 1970s until 2021. His PhD thesis was completed in 1967 [13]. His last major synthesis — on ciliary evolution and the root of the eukaryote tree — was published shortly after his death in 2022 [12]. His career has been devoted to making a deep synthesis of the evolution and diversity of cells.
Tom remained a true naturalist and generalist throughout his life while being a meticulous specialist and polymath at the same time. His main research aim was to put all major groups of organisms in their proper phylogenetic place and understand the transformations between them. He developed and constantly updated his taxonomic system encompassing all cellular life and built evolutionary scenarios with utmost molecular and mechanistic detail.
His evolutionary theories and classifications developed and improved over the years. He constantly incorporated new molecular and cellular evidence or newly discovered lineages of organisms. His theories and classification system also showed a rich evolutionary history, just like organismal evolution they depicted. In his work, he always weighed up and searched for congruence among multiple lines of evidence. He was equally well versed in cell biology, cell behaviour, microscopic structure, molecular function, sequence phylogeny, and the fossil record.
His thinking about major steps in evolution, or megaevolution, was inspired among others by the palaeontologist and evolutionist George Gaylord Simpson [14]. Tom argued like Simpson that the origin of the higher ranked groups of organisms involved more fundamental, rapid and concerted changes — termed quantum evolution by Simpson — “not just little ones as in diversification of Darwin’s finches” [15]. This grand vision of evolution permeates his work and is evident equally in his taxonomic and cell-evolutionary theories.
Understanding the origin and evolution of cells
Tom’s work encompassed a broad range of topics in cell evolution and taxonomy. He started his career working on the electron microscopy investigation of the Chlamydomonas cell cycle and developed a sharp eye to functionally and conceptually interpret ultrastructural data.
Since the PhD work, a central theme of his interest was the origin of complex eukaryotic cells during the process called eukaryogenesis (the term goes back at least to F. J. R. Taylor, 1976 [16]). He also worked on the origin of the first cells, the origin of archaebacteria from eubacteria, the evolution of prokaryotic cell structure, and many other evolutionary transitions. Tom also wrote extensively on symbiosis, including the origin of mitochondria, chloroplasts and secondary endosymbioses and made major conceptual and taxonomic contributions.
His cell evolutionary theories were detailed, specific and testable with clear predictions. His theory on the origin of the first cells features an intermediate stage called ‘inside out-cell’ or obcell [17]. His theory of eukaryogenesis — called the neomuran theory — describes a complex series of cellular transformations from a eubacterium to archaebacteria (archaea) and eukaryotes (the sister group to archaea in Tom’s theory). Other research topics included genome evolution, the origin of telomeres, introns and the evolution of genome size.
Tom also worked out his own and ever-evolving systematics of the entire tree of cellular life. He also described about 150 new protist species and classified them together with countless other species and lineages.
Major transitions in cell evolution: Eukaryogenesis
A major recurrent theme throughout Tom’s research has been the transition from prokaryotes to eukaryotes representing the “single greatest cellular upheaval in the history of life”.
He was dissatisfied by the explanations of Lynn Margulis [18] that symbiosis could alone explain the origin of eukaryotes. In his view the exclusive focus on symbiosis distracted attention from the problem of how the whole genetic system, the cell cycle, membrane organisation, sexual cell fusion and all the other structures evolved from the prokaryotic to the eukaryotic pattern.
It was nevertheless the publication in 1970 of Margulis’s book on the origin of eukaryotes [19] and reviews by Stanier [20] and Echlin [21] that led him to start thinking about eukaryotic organelle evolution.
In the early 1970s, he started to seek a coherent explanation for the drastic changes in cell structure and organisation during eukaryogenesis. His explanations tried to integrate evidence from cell ultrastructure, molecular cell biology, phylogeny and the fossil record.
In 1975 Tom set out to explain the basic logic of his theory on eukaryogenesis [22]. The paper involved several interdependent ideas tied together to give a comprehensive picture. This early work still proposed an autogenous model for mitochondria (and chloroplasts), ideas that Tom quickly abandoned after enough molecular evidence pointed to their symbiogenetic origin.
Tom’s theory of eukaryogenesis is known as the neomuran theory [23]. The key features of the theory include the derivation of both eukaryotes and archaebacteria (that Tom argued were sister groups) from an actinobacterium [24] or in the final version of the theory a Planctobacterium [7].
In Tom’s theory, the cytoskeleton and phagotrophy made the eukaryotic cell [25]. The first step in this transformation was the loss of the murein cell wall (and loss of the outer membrane in the plactobacterial version) followed by stabilising changes. In the archaeal lineage, adaptation to thermophyly led to the change in membrane and cell-wall chemistry. In the eukaryotic lineage, the evolution of a dynamic cytoskeleton, phagocytosis and intracellular prey digestion were the first decisive evolutionary innovation, triggering a major upheaval of cell structure.
Tom remained critical of the serial endosymbiosis theory of Lynn Margulis that put a major emphasised on symbiosis without explaining how this led to other major cellular transformations. Margulis’ scenario on the symbiotic origin of cilia [26] was also dismissed because of the lack of evidence or testable predictions. Tom argued that a comprehensive theory should explain the origin of 60 unique eukaryotic properties [27] (almost all still valid). He was equally critical of syntrophic [28] or mitochondria-first theories (e.g. [29]) based on mechanistic and cell-biological grounds.
Tom’s theories always emphasised coevolution between different parts of the cell and invoked concerted changes to explain major changes in cell structure. For example, the loss of the prokaryotic cell wall was a prerequisite to evolve a flexible membrane enabling phagotrophy. Phagotrophy then destabilised the ribosome attachment sites that needed to be rescued by internalising them, giving rise to the endoplasmic reticulum. Phagotrophy also enabled the later acquisition of mitochondria, rather than mitochondria being the primary driver of eukaryogenesis [30].
He lamented that it was “perverse not to recognise that phagotrophy was the biggest discontinuity in nutritional mode in the history of life that completely changed the scaling laws for both nutrition and genome size evolution with cell volume.” [7]
Following the origin of mitochondria that escaped digestion after being phagocytosed, self-splicing introns derived from the mitochondrial ancestor giving rise to eukaryotic introns [11]. These introns then were able to spread “like wildfire” after the nuclear envelope evolved. Major changes in genome organisation also enabled the evolution of much larger genome sizes of eukaryotes than of prokaryotes. Most fundamental in Tom’s view was an indefinite number of replicons to initiate DNA replication simultaneously — impossible in prokaryotes.
He also provided the first explicit molecular model of how eukaryote chromosome ends (telomeres) might be replicated and could have originated in evolution [31]. “The primary reason d’être for telomeres was to solve the end replication problem of linear chromosomes” [32]. His model still applies to some viruses, although he could not foresee the discovery of telomerase.
His explanations have developed with ever growing detail from 1975 to 2020 as steadily growing evidence was incorporated. He emphasised the need to study this transition from an intracellular coevolutionary perspective. Tom was probably the only one who explained in detail how each of the 60 major eukaryotic innovations evolved and how each related to the others. His molecularly and mechanistically exquisite evolutionary scenarios continue to provide a rich source of inspiration. Tom was well aware that the latest phylogenies by others showed that eukaryotes derived from within Archaea. He strongly argued that these were artefacts [7]. He thought that the unique membrane characteristics of Archaea were an important synapomorphy that argued for an archaeal clade to the exclusion of eukaryotes. Regardless of the phylogeny, many arguments and cellular details of Tom’s eukaryogenesis theory including his models on the evolution of the nucleus, endomembranes, cytoskeleton, cilia, introns, etc. remain valid.
The Archezoa theory
The Archezoa theory represents a key stage in the development of Tom’s ideas on eukaryogenesis [23,33,34].
Archezoa was established by Tom as a putative, primitively amitochondrial taxon with cells also lacking peroxisomes and Golgi dictyosomes. The theory had three logically independent hypothesis: i) members of Archezoa represent a basal, paraphyletic taxon; ii) most or all members of Archezoa are primitively amitochondriate and lack peroxisomes; and iii) such amitochondriate phagotrophic protists represent a key stage in the evolution of eukaryotes before mitochondrial uptake.
In the phylogenetic hypothesis, Archezoa were a paraphyletic group that included the metamonads like Giardia lamblia, the parasitic microsporidia, the amitochondrial amoebae Archamoebae and Parabasalia (trichomonads and hypermastigote flagellates).
Early molecular support for members of Archezoa being early-branching with potential ancestral features came from the analysis of rDNAs. In 1986 Vossbrinck and Woese sequenced the rDNA of the amitochondriate microsporidian Vairimorpha necatrix that was simpler than in other eukaryotes [34,35]. Later, Sogin’s 16S-like rRNA sequence tree (including both 18S (eukaryotes) and 16S (prokaryotes) sequences) suggested that apparently mitochondria-less microsporidia and Giardia lamblia were early-diverging eukaryotes [36].
The original phylogeneitc hypothesis for Archezoa [33] was rapidly revised in subsequent papers, based on new data. Parabasalia were the first group to be removed from the Archezoa because rDNA trees suggested that they derived from mitochondrial ancestors [37].
Later, Microsporidia were also removed from the Archezoa [38] because protein sequence data suggested that they were highly degenerate fungi and a nuclear mitochondrion-derived Hsp70 sequence showed that they once contained mitochondria [39–41]. Their deep branching in Sogin’s tree was due to long-branch artefact. The phylogenetic evidence that Entamoeba [42], Phreatamoeba [43], and Pelomyxa [44] are all secondarily amitochondrial also prompted Tom to remove Archamoebae from the Archezoa [38,45].
The second hypothesis on the primitive amitochondriate character of Archezoa was also revised. Importantly, Tom had always explicitly recognized the possibility that some of the archezoa might have evolved by the secondary loss of mitochondria [33].
“Though it is possible, as traditionally assumed, that some or even all of the Archezoa originated from mitochondrion-containing protozoa by the loss of mitochondria, there is no evidence that this is true for any of them.”
Based on the new phylogenetic data, Tom early on suggested that the double-membraned hydrogenosomes in Parabasalia evolved secondarily from mitochondria [46]. This was later confirmed molecularly by the analysis of the mitochondrial chaperonin 60 of Trichomonas vaginalis [39,47,48].
By the late 1990s, it became clear that i) there are likely no primitively amitochondriate eukaryotes and ii) that members of the original Archezoa were not deeply-branching lineages in the eukaryote tree [49,50].
The archezoa theory also postulated a phagotrophic ancestry of eukaryotes as its third hypothesis. This posited a primitively amitochondriate stage in eukaryogenesis with cells already having a dynamic cytoskeleton, endomembranes, a nucleus and the ability to engulf other cells. According to this hypothesis phagotrophy was essential for mitochondrial endosymbiosis (‘phagotrophy first’). This cell-evolutionary hypothesis was logically independent from the two phylogenetic hypothesis.
“However, the reasons for postulating that the α-proteobacterium was taken up by an early amitochondrial and non-chimaeric eukaryote after the origin of the cytoskeleton, endomembrane system and phagocytosis [] remain compelling, even if no primitively amitochondrial descendants of this early non-chimaeric phase of eukaryote evolution remain” [33,38,51]
Tom remarked that critics purposefully confounded the different elements of his archezoa theory (the ‘demise of Archezoa’) with the aim to dismiss phagotrophy-first models based on the superseded phylogenetic hypotheses.
The phagotrophy theory has been further elaborated in later papers as a separate evolutionary hypothesis [24,52] (see previous section) and remains a prominent and valid theory today (e.g. [53]).
Primary and secondary endosymbioses
While Tom dismissed symbiosis-only or symbiosis-first explanations for eukaryogenesis, he knew how essential symbiogenesis was in cell evolution and the origin of major groups of eukaryotes. Ever since his PhD work, he made major contributions to understanding the role of symbiosis and organellar heredity. In 1967, he described the fusion of chloroplasts during the Chlamydomonas cell cycle.
In his later work, he frequently emphasised that symbiogenesis involves radical alteration and merger of two organisms and is very different from symbiosis. Organelles thus are not just “bacteria living in eukaryotes” [19], “but novel, chimeric organelles resulting from intimate merger at all scales of two fundamentally different organisms” [54].
He early on recognised that the key distinction between an intracellular symbiont and an endosymbiotic organelle is the presence of a “generalized protein-import mechanism for importing any protein to which the correct topogenic sequence is added” [54,55].
Based on the complexity and evolutionary difficulty to evolve such mechanisms for protein targeting Tom argued that symbiogenesis occurred only very rarely in the history of life.
“Symbiogenesis is rare because it requires not just the addition of an extra genome to a preexisting cell, but also the integration of host and symbiont genetic membranes by the evolution of novel protein-targeting systems” [56]
The prevailing view in the early 1980s was that chloroplasts in various algal groups derived from several independent endosymbiotic events. In 1982 Tom suggested instead that only two endosymbioses were needed to account for the origin of all plastids [57].
One led to the origin of green plants, red and glaucophyte algae that all derived from a single primary endosymbiotic event during which a biflagellate cell engulfed a cyanobacterium. This is now universally accepted, and is evidenced by e.g. the similarities in the mechanism of protein targeting to primary chloroplasts and molecular phylogenies [58,59].
Tom extensively discussed in his writings the stages of primary endosymbiosis including key steps such as the evolution of membrane transporters for exporting photosynthetic products and machinery for importing nuclear-coded proteins into chloroplasts [54,56,57].
Looking at another group of eukaryotic photosynthetic organisms, where the plastids are located within a periplastid membrane in the lumen of the rough ER, he proposed that all of these eukaryotes (and their non-phototsynthetic relatives) evolved after a distinct endosymbiotic event, during which an aplastidic eukaryote engulfed a primary-plastid-bearing eukaryote.
These considerations led Tom to establish the predominantly photosynthetic kingdom Chromista (regnum novum)[60]. This group was distinguished from the predominantly phagotrophic Protozoa and was united by a plastid in the ER membrane. The Chromista initially included Cryptophyta and Chromophyta [60] and later also the mainly photosynthetic groups Chlorarachniophyta, Cryptista, Heterokonta, and Haptophyta and their non-photosynthetic relatives [9].
A distinctive feature of chromist chloroplasts is that they are within the lumen of the rough endoplasmic reticulum (RER) and in addition are surrounded by a smooth periplastid membrane. The periplastid membrane derived from the plasma membrane of the eukaryotic photosynthetic symbiont that was engulfed by the host. The symbiont then entered the RER lumen by fusion of the phagosome membrane with the nuclear envelope [9]. This secondary endosymbiotic event thus led to a chloroplast (derived from a red alga [61] ) with four membranes representing the most complex membrane topology and protein targeting mechanisms among eukaryotes. The primary reason for suggesting a single secondary symbiotic event at the base of the Chromista was “to minimise the number of separate origin of novel protein import machinery in secondary symbiogenesis”.
The Chromista was an influential phylogenetic hypothesis that Tom and others continued to test. Sequence analyses of cryptomonad 18S rDNA yielded unstable molecular trees leaving Tom undecided about the monophyly of chromists [61].
“This probable polyphyly of the chromophyte algae, if confirmed, would make it desirable to treat Cryptita, Heterokonta, and Haptophyta as separate kingdoms, rather than to group them together in the single Chromista.” [61]
Chlorarachniophyta were later removed from the chromists because sequence evidence demonstrated their independent symbiogenetic origin (in this case a green alga engulfed by a non-photosynthetic host) [62].
Around the same time, chromists were extended when Tom suggested that the last common ancestor of Alveolata (a group that includes protists like ciliates) was photosynthetic and its plastids originated by the same secondary endosymbiogenetic event. Chromista and Alveolata together form the chromalveolates [62]. The groupings continued to evolve as new data were incorporated and the last synthetic work on chromists was published in 2017 [63].
Another important direction in Tom’s research on chromists involved membrane targeting and the evolution of nucleomorphs. Nucleomorphs are reduced eukaryotic nuclei that are remnants of the engulfed red or green algal symbiont present in cryptomonads (red algal origin) and chlorarachneans (green algal origin). Decisive evidence for the nature of these ancient eukaryote-eukaryote cellular chimaeras came from sequencing nucleomorph genomes. In collaboration with Uwe Maier and Susan Douglas Tom fully sequenced the reduced nucleomorph genome of the cryptomonad Guillardia theta proving that this and related cells are indeed such ancient chimaeras [64,65]
The chromist theory is an influential synthetic theory that emphasizes the importance of cell ultrastructure, protein targeting, phylogeny and symbiogenesis. As Tom always argued, phylogenies alone are not sufficient to understand and reconstruct cell evolution, due to systematic biases, rapid radiations making it hard to resolve all clades and problems related to paralogy and gene exchange derived from endosymbiotic gene transfers.
Membrane heredity:
Nucleomorph membranes were in evolutionary continuity with the nuclear membrane of the once free-living eukaryotic algae. Likewise, the periplastid membrane derived from the plasma membrane of these once free-living cells that has been maintained over hundreds of millions of years through subsequent divisions.
These considerations were generalised by Tom within the overarching concept of membrane heredity [56]. Membranes are inheritence systems, that give rise to new membranes through growth and division and maintain their identity through autocatalytic processes.
“Two universal constituents of cells never form de novo: chromosomes and membranes. Unlike ribosomes and microtubules, which form by self assembly, cell membranes always form by growth and division or fusion of pre-existing membranes. The diverse membranes of the millions of extant species are all lineal descendants of those of the first bacterial cell.” [56]
The theory distingushes genetic membranes from other membranes. Genetic membranes are those membranes that cannot form de novo. He illustrated this concept with a thought experiment. If a cell lost all membranes of a certain type (e.g. all mitochondria lost their outer membranes), would there be a mechanism to regrow them? There is no such mechanism for the plasma membrane, the ER, the mitochondrial and chloroplast inner and outer membranes and other membrane derived by secondary symbiosis. This is in contrast to non-genetic membranes that can grow by budding from preexisting membranes such as the Golgi or endocytic vesicles. The identity of genetic membranes is maintained by autocatalytic processes, which include membrane proteins forming type-specific insertion, transfer or retention machineries [56].
Genetic membranes thus have special evolutionary importance as inheritance systems. The history of genetic membranes was always central in Tom’s cell evolutionary theories. For example, only a proper understanding of membrane heredity can explain why chlorarachnean algae have four chloroplast membranes. These chloroplasts evolved by secondary endosymbiosis, which means the symbiosis of an eukaryote host with a phototsynthetic eukaryotic alga (a green alga in the case of chlorarachnean algae), and retained the two chloroplast membranes of the cyanobacterium derived from a primary endosymbiotic event, a periplast membrane derived from the former plasma membrane of the eukaryotic alga that was engulfed and the RER of the host cell [56].
“The importance of membrane continuity in evolution, and of the idea of membrane heredity and the protein-targeting mechanisms that mediate it, is particularly evident in the symbiogenetic origins of eukaryotic organelles”
Tom argued first that the outer mitochondrial membrane derived from the outer membrane of the gram-negative bacterial symbiont [33] (now universally accepted) contrary to Lynn Margulis’s 1970 version which imagined it came from the host food vacuole membrane.
The concept of membrane inheritence is a good example of Tom’s holistic thinking about cells. He conceptualised cells as systems of mutualistic symbiosis between genes, membranes, enzymes and skeletal elements, rather than entities ruled by a “unilateral informational dictatorship by DNA like a divine nuclear authority as it is too often misrepresented”.[15]
Timeline of evolution and fossil evidence
Tom always placed great emphasis on establishing a rigorous timeline for major events in cell evolution. His first paper on the topic, published in a palaeontology journal [10] was pioneering in its attempt to critically integrate palaeontology with studies of extant cells. He argued that the first cells date from 3500 My ago (Mya) and were likely anaerobic, photosynthetic Negibacteria and that the last universal common ancestor (LUCA) must have been a complex prokaryote with at least 1000 genes and cannot possibly have been a feature-less ‘progenote’ as proposed by Carl Woese. He also repeatedly argued that no fossil older than 900 Mya can be assigned with certainty to any group within crown eukaryotes. [15]
Tom also repeatedly warned against the uncritical adoption of molecular clocks, including the “oxymoronic ‘relaxed clock’ computer programs” [66]. He fumed “that meaningless ‘temporal’ analysis of paralogue trees that so dramatically flout oversimplified assumptions of ‘clock’ algorithms—useful only if applied to relatively uniformly evolving single orthologues and calibrated by fossil dates needing no signifcant extrapolation beyond the direct evidence.” [7].
There is increasing analytical evidence supporting Tom’s view: molecular clocks show puzzling mismatches between the estimated ages and the fossil record and tend to overestimate the ages for large clades due to systematic biases and artifacts [67,68]. Calibrations are also often problematic, “driven by palaeontologists’ ‘my fossil is older than yours’ competition” [66].
Tom was well aware of these limitations and of the extremely accelerated rates that can happen in the stem lineage of some clades. Although his treatment of these topics was not mathematical, his thinking on the ‘tempo and more of evolution’ followed the ideas described by Simpson (whom Tom regarded very highly) in his famous book [14]. Recent mathematical analyses provide a quantitative framework for what was obvious to Simpson and also Tom, namely that large clades may originate by explosive early radiations during which molecular evolution can occur at elevated rates [69]. For example, Tom argued that the extremely rapid early radiation of chromists groups could explain why it is hard to recover all chromists in molecular trees or why there is no evidence for the early differential sorting of gene duplicates [63].
This was a recurrent theme in Tom’s writings, in particular he often lamented about the persistence of the “regrettable tendency to take 16S rRNA trees as gospel truth and ignore other evidence” [70].
Evolution of basal bodies, cilia, and the transition zone
Ever since his PhD work on Chlamydomonas, Tom has developed detailed scenarios and wrote comprehensive papers like none other on the evolution of eukaryotic cilia. Cilia — eukaryotic organelles consisting of a membrane-bound motile ciliary shaft that connects via a transition zone to a supporting basal body — represent outstandingly valuable characters for cell evolution. In particular, the transition zone is highly variable amongst but conserved within most major eukaryote lineages making it a useful marker of true evolutionary affinities.
The last paper of Tom’s — published after his death — deals in excruciating detail with the evolution of the transition zone [12]. This 100-page-long magisterial mega-paper is a good example of his writing style and synthetic approach to cell evolution. The paper not only discusses transition zone diversity and evolution across eukaryotes, re-analysing every single electron micrograph ever published, he also provides a grand synthesis of the evolution of the structure, its relation to ciliary motility and feeding. By mapping character states across all lineages, Tom also aims to clarify the evolutionary affinity of several lineages and root the eukaryotic tree. The paper has 26 major conclusions about eukaryotic phylogeny, taxonomy and cilium evolution. This is not uncommon in the lifetime opus of Tom but unique and unprecedented in the literature.
The unique scholarly merit of Tom’s approach is the synthesis of information from many different sources, including ultrastructure, molecular phylogeny, swimming mode and molecular cell biology. This allows Tom to propose a rich set of very clear and specific functional hypotheses on ciliary function and evolution. He also establishes large patterns of cilium and locomotory evolution across eukaryotes.
It is impossible to give credit to all the detailed insights and hypotheses in this body of work. Testing all the immense number and richness of specific ideas, mechanistic and evolutionary hypotheses in this single paper alone (considered one of the most important ones by Tom in his entire career) would take a decade of work in cell biophysics, ultrastructure and comparative genomics.
Animal evolution
Tom also had an interest in the evolution of multicellularity, in particular the origin of animals (Metazoa). He made seminal contributions to the phylogeny of protists related to animals, to mitochondrial genome evolution in corals and to conceptualising animal origins.
His phylogenies contributed to early molecular evidence that the choanoflagellates were the sister group to the monophyletic animals [8,9,61,71]. Based on the morphology of their collared-cells, a close relationship of choanoflagellates to sponges has already been suggested by Henry James-Clark [72] and W. Saville Kent [73].
Tom started to elaborate his ideas about animal phylogeny and transitions in the 1980s. At a Systematics Association Symposium on Lower Invertebrate Origins in 1984 he presented these bold ideas. When he was challenged on how to test them, he replied that by sequencing mitochondrial genomes from various animals. For the only time in his career “almost the entire audience burst out laughing, presumably at the very idea that sequence information could be relevant to phylogeny”. The audience consisted of “classical comparative zoologists with no molecular or cell biological perspective, sprinkled also with dogmatic Germanic Hennigian cladists” [15].
After this symposium, Tom resolved to learn how to isolate and sequence mitochondrial DNA and do molecular phylogeny. He started with cnidarians, sponges and choanoflagellates during his sabbattical in Australia. His work on cnidarian mitochondrial genomics initiated by his 1987 cloning of the Sarcophyton mitochondrial genome [74]. The genome had a MutS gene encoding an excision repair enzyme which had never been found before in any mtDNA. Together with Des Clark-Walker, Tom argued that this evidenced evolutionary gene transfer from the nucleus to mitochondria, the first such example [75,76].
Tom also worked on other animal groups, including the molecular phylogeny of sponges [77] and lophophorates [78,79], a group including animals with a filter-feeding lophophore (e.g. bryozoans and brachiopods).
The broader phylogeny of animals was also enriched by Tom’s classification of related protists, including Corallochytrium [80]. In 1987 Tom suggested the clade Opisthokonta to include animals, fungi and choanoflagellates unified by a single posterior cilium in the unicellular motile stages (in chytrids within fungi) [81]. The relationship of fungi, animals [71,82,83], and also choanoflagellates [8,9,61,71] was later confirmed by molecular phylogenies.
In his theoretical work, Tom presented detailed and well-argued transition scenarios [66]. He envisioneged a filter-feeding benthic stem metazoan formed by the aggregation and settlement of choanoflagellate-like cells. “I contend that it was not the presence of potential glue molecules, but the rare ability of choanoflagellate cells to stick together yet still feed as before that made stem choanoflagellates our ancestors.” [66]
Tom’s idea on the origin of nervous systems were also original and compellingly presented [66,77]. “Synapses evolved to make ciliated larval settlement faster and more effective by neural coordination of concerted banks of nematocysts under the control of ciliated sensors that selected the best sites.”[66] While he also critically evaluated the molecular evidence for animal origins, he did not think highly of molecule-centric approaches. On the widespread practice of homologysing body reagions across phyla based on similarities in gene expression programs he remarked: “Thinking human and grasshopper heads structurally homologous is as bad as calling a vacuum cleaner and light bulb homologous, because identical switches can turn both on.”
Building a consistent phylogenetic and macroevolutionary system
In 1969 Tom red a review by Whittaker that presented a five-kingdom classification system of the whole living world [84]. This was of great interest to him as he soon realised that the classical division into animal and plant kingdoms was inadequate to represent microbial life. He resolved to develop a highest level classification of life based on sound ultrastructural phenotype, cell function and phylogeny.
Tom was a practitioner of evolutionary taxonomy, which takes into account both branching order and the magnitude of phenotypic differences (unlike cladistic taxonomy, which only considers branching order). This system clearly separates phylogeny —reconstruct the branching pattern of life — and classification —place organisms into monophyletic taxa based on the branching pattern and the magnitude of phenotypic differences — in his view often confused nowadays. Evolutionary taxonomists sometimes need to introduce paraphyletic taxa, because “that is how evolution works” [70].
“Categories like kingdom and phylum refer to taxa, which should always be monophyletic (i.e. holophyletic or paraphyletic), and never polyphyletic, not to clades or grades; thus many taxa are clades but some are necessarily paraphyletic grades because that is how evolution works.”[70]
The “how evolution works” of course refers to ideas related to Simpson’s quantum evolution, driven by the invasion of previously unexploited major adaptive zones. This leads to rapid divergence and transformative changes while at the same time members of the ancestral group can remain unchanged. This mode of unequal evolution is recapitulated by model simulations [69]. Therefore it is not only branching that matters in evolution, as in the Hennigian view, argues Tom, but also the ‘tempo and mode’ [14].
In cladistic phylogeny mammals are nested within fish and are classified as fish (Teleostomi); in evolutionary taxonomy one has to say that mammals derived from, not nested within, fish (and thus are not classified as fish).
Tom argued that Willi Hennig destroyed the original purpose of the term monophyly (clades descended from a shared common ancestor) by using it for holophyly alone (to include all descendants of that ancestor) but exluding paraphyly (monophyletic group with the exclusion of a subgroup). This has sown “immense confusion (two conflicting meanings now for monophyly) that still haunts us” [70].
Elsewhere he remarked that not caring about rank and opposing Linnean categories above family is “a minority prejudice but influential in some circles which makes it harder to communicate higher classification effectively”.
Tom’s classifications were regularly updated and he did not shy away from creating new high-level taxa at every revision of his system. He vigorously defended his principles. “To those who believe that classifications should not be based on hypotheses, I simply say that all phylogenetic classifications are based on similar sorts of phylogenetic hypotheses. The degree to which a phylogenetic hypothesis needs to be corroborated before being used as a partial basis for a classification is a matter for scientific judgement by individual systematists. It is neither philosophically correct nor good manners to call the phylogenetic hypotheses of others speculation, and to treat one’s own as accepted facts.” [38]
Based on these principles, Tom created his own classification system that he constantly revised in light of new evidence. Given the rapid increase in the number of sequences and the number of lineages discovered — to a large part also due to his research work — this system evolved rapidly. His new high-level taxa were often not included in consensus phylogenies. The speed of changes in his taxonomic system may have been too fast to leave time for consensus building and testing. He was criticised for formalising names prematurely and reorganising classifications single-handedly. His global taxonomic system thus stand as its own expanding universe. Nevertheless, a very large number of taxa he proposed are now part of the current consensus classification including Opisthokonta, Ascomycota, Ichthyosporea, Alveolata, Rhizaria, Haptista, Jakobida, Euglenozoa and many more [85].
Tom also proposed and updated a “formal nomenclaturally valid treatment of bacteria higher taxonomy” [7,24]. He also described over 150 new protozoan and chromist species from all continents except Antarctica.
Rooting the tree of eukaryotes and the tree of life
In phylogeny, the direction of evolutionary change can only be interpreted correctly if the tree is rooted on an outgroup. This rooting problem is technically challenging and prone to various artefacts. Tom dedicated a large number of papers to the problems of rooting in various groups, in particular to find the root of the eukaryotic tree and of the tree of all life.
Rather than relying on sequence trees only, his approach to rooting was also integrative by collating evidence from molecular trees, morphology, and the fossil record. He then weighted the evidence and critically combined it with transition analysis.
To root the eukaryotic tree, he introduced the use of derived gene fusions, representing rare genomic events. The resulting ‘unikont-bikont’ rooting was an important framework for many years [86]. In his last paper, he proposed a new rooting, between the malawimonads — small biciliated protists — and the rest of eukaryotes. This was based on the very simple and asymmetric ciliary transition plate in malawimonads that he proposed may represent the ancestral condition for all eukaryotes [12]. Tom’s malawimonad root differs from the currently favoured ‘excavate-like’ ancestry of eukaryotes [87] hinting at the limitations of morphology-based rooting.
A lot harder and still unresolved problem is to root the tree of life. In phylogeny, rooting a group of organisms relies on comparisons with outgroups. However, as so nicely put by Tom, “outgroups for the entire tree are air, rocks and water, not other organisms, vastly increasing the problem” [70]. But solving this problem is essential to understand the nature of the most ancestral cells and thus the origin of life.
To pinpoint a root, Tom used transition analysis, a method to determine the likely direction of evolutionary changes (e.g. bird wings evolved from forelimbs used for walking, not the opposite). Such analyses can in principle polarize change unambiguously in the absence of a fossil record or outgroups. He listed a large number of characters that polarised evolution and supported a root for the universal tree within eubacteria [70]. Given that rooting the tree of life is the most difficult of all phylogenetic problems and that the evidence for the widely accepted root between archaea and bacteria [88] may be due to phylogenetic artefacts [70], Tom’s hypothesis for a bacterial root remains to be tested.
Legacy
It is impossible just yet to give a final picture of Tom’s legacy and influences. We will have to keep re-reading his immensely rich papers and continue to learn from them and be inspired by them. They contain so many clear and testable evolutionary and mechanistic hypotheses to work out in detail.
The first important legacy that already emerges is that we should embrace his integrative, synthetic method of approaching cell evolution. He argued that evolutionary hypotheses “should be explicit and detailed to facilitate reasoned criticism, refutation and improvement; not non-committal and vague to evade referee objections which can make them untestable and scientifically useless.” [89]
As an example of a vague theory, Tom repeatedly criticised as “superficial” the idea of the ‘progenote’ introduced by Carl Woese for the last universal common ancestor. Tom pointed out that the “‘progenote’ was an idea empty of content and thus untestable and unimprovable; its only specification was that replication, transcription and translation were poorly developed and inaccurate.” Accordingly, the progenote idea can not be challenged on specifics.
One reason why there is a “dearth of coherent imaginative but critical synthesis as done by Darwin and Simpson” may be that when it is attempted, it is “often harmfully dismissed as speculation and deterred by journal publishing and refereeing practices”. (One positive recent change is that eLife now allows authors to include an “Ideas and Speculation” section in their papers https://elifesciences.org/inside-elife/e3e52a93/elife-latest-including-ideas-and-speculation-in-elife-papers. I am sure Tom would have welcomed this policy.)
Second, Tom was famous for frequently changing his view in quite important ways in the light of new data, an ability we should try to emulate. As he wrote in his autobiography “It is better to discard one’s intellectual mistakes once they are clearly disproved than to die with them out of unscientific hankering for consistency.”
Third, Tom was fearless in debate and criticism, — maybe his most controversial character trait — which in his writing may have come across harsher than in person. His longer papers famously single-handedly picked apart dozens of other papers and theories with incisive criticism, editorializing on the shortcomings of others’ points of view. He was merciless at times: “to suggest that a bacterium with a normal wall engulfed another cell ‘by an unspecified mechanism’ is magic, not science.”[24]. I know how authors could have felt after such criticism. A bit more ‘Britishness’ could have softened some of this, maybe engaging more people in debate rather than elicit fight-or-flight. I also received my fair share in public peer reviews (but stood my ground). But what then remains a puzzle for me, is how different Tom was in person. I had the pleasure of visiting Ema and Tom in their home in Oxford to record a long interview (rather monologue) with Tom. He was exceedingly kind and thoughtful, weighing alternative views and discussing my critical comments. I could not have foreseen such kindness based on the papers and our written exchanges. He could be tougher with others, though. I remember how Tom debated Stephen Jay Gould about species-level selection during the EMBL PhD Symposium “Evolution” in 2001. Gould gave the example of the naked mole rat but Tom continued to press. Gould then simply dismissed him, probably due to a combination of pride and lack of arguments. It was memorable. Others, irritatingly, also just did not bother to engage with him [70]. But debate is essential for scientific progress. So we may need to go back to Tom’s papers and learn how to deliver criticism of an entire field and then learn to respond to it.
And finally, Tom was an exceptional and eccentric writer and his prose often has also literary aesthetics and value. He could pencil mesmerising and memorable paragraphs, like this opening one from his final treatise of chromists:
“Only animals and bacteria have more phyla than chromists, but even they cannot match chromists in their remarkable range of contrasting adaptive zones — from giant brown algal kelps longer than a blue whale to ciliates like Paramecium, dinoflagellates that power coral reefs or kill shellfish, the most abundant predators in soil (sarcomonad Cercozoa), parasites like Toxoplasma whose cysts are allegedly lodged in a third of human brains and Plasmodium that causes malaria, diatoms whose silica frustules were once essential for making dynamite or polishing astronomical telescope mirrors, and foraminifera or haptophyte plankton like Emiliania that can be seen from outer space and made the white cliffs of Dover with their calcareous scales and are probably the most speciose photosynthetic oceanic flagellates and exude volatile chemicals that affect cloud formation and global energy balance.” [63]
Or on cellular mergers via symbiogenesis: “Transcending mutation within one lineage, symbiogenesis hugely shaped modern life. Cellular chimeras had rare evolutionary roles: Though chimeric multicells (e.g., graft hybrids) cannot persist through the germline to make centaurs or Pegasus, Empedocles’s evolutionary mechanism sometimes works for unicells.”
Volumes could be filled with similar quotes by Tom. He was like Pegasus, bringing lightning and thunder from Olympus.
Awards and Recognition
Fellow of the Linnean Society of London (1980)
Fellow of the Institute of Biology (1983)
Fellow of the Royal Society of Arts (1987)
Fellow of the Canadian Institute for Advanced Research (1988)
Fellow of the Royal Society of Canada (1997)
Fellow of the Royal Society of London (1998)
International Prize for Biology from the Emperor of Japan (2004)
Linnean Medal for Zoology (2007)
Frink Medal of the Zoological Society of London (2007)
Brief Author Profile
Gáspár Jékely studied Biology and obtained his PhD in 1999 at the Eötvös Loránd Universities in Budapest. He then worked at the European Molecular Biology in Heidelberg as a postdoctoral fellow. Between 2007-2017 he was a group leader at the Max Planck Institute for Developmental Biology in Tübingen, Germany. In 2017, he moved to the Living Systems Institute at the University of Exeter as Professor of Neuroscience. Since 2023, he is Professor of Molecular Organismal Biology at Heidelberg University where he works at the Centre of Organismal Studies. His research interests include the structure, function and evolution of neural circuits in marine ciliated larvae and the origin and early evolution of cells and nervous systems. He has met Thomas Cavalier-Smith in 2001 during a conference on Evolution at EMBL and has closely followed his work. Later, he invited Tom to the Max Planck Institute in Tübingen. In 2014, the author visited Tom and his wife Ema in their home in Oxford and recorded an interview with Tom.
Acknowledgements
In are very grateful to Ema Cavalier-Smith for sharing the unpublished autobiography of Tom and for providing personal details and feedback on the manuscript. I also thank… for their suggestions on an earlier version of this memoir. The frontispiece portrait photograph was taken in x by ..
