The Tonian Period (ca. 1000–720 Ma) follows the ‘boring billion' in the Mesoproterozoic Era and precedes ‘snowball Earth' glaciations in the Cryogenian Period. It represents a critical transition in Earth history. Geochemical data indicate that the Tonian Period may have witnessed a significant increase in atmospheric pO2 levels and a major transition from predominantly sulfidic to ferruginous mid-depth seawaters. Molecular clock estimates suggest that early animals may have diverged in the Tonian Period, raising the intriguing possibility of coupled environmental changes and evolutionary innovations. The co-evolution of life and its environment during the Tonian Period can be tested against the fossil record by examining diversity trends in the Proterozoic and evolutionary innovations in the Tonian. Compilations of Proterozoic microfossils and macrofossils apparently support a Tonian increase in global taxonomic diversity and morphological range relative to the Mesoproterozoic Era, although this is not reflected in assemblage-level diversity patterns. The fossil record suggests that major eukaryote groups (including Opisthokonta, Amoebozoa, Plantae, and SAR) may have diverged and important evolutionary innovations (e.g. multicellularity and cell differentiation in several groups, eukaryovory, eukaryote biomineralization, and heterocystous cyanobacteria) may have arisen by the Tonian Period, but thus far no convincing animal fossils have been found in the Tonian. Tonian paleontology is still in its nascent stage, and it offers many opportunities to explore Earth-life evolution in this critical geological period.
A pressing need to improve the Tonian fossil record
Geologist Roger Buick  wrote more than 20 years ago that ‘the dullest time in Earth's history seems to have been the Mesoproterozoic, the era between 1600 and 1000 Ma ago'. The Mesoproterozoic Era has since been known as the ‘boring billion'—a period of apparent quiescence as far as the global carbon cycle is concerned , although important evolutionary events did occur in the Mesoproterozoic [2–6]. In the grand scheme of Earth history, the Tonian Period follows the so-called ‘boring billion' of the Mesoproterozoic Era  and precedes the Cryogenian ‘snowball Earth' glaciations each lasting millions of years . Because of its critical location in the geological timescale, the Tonian Period holds the key to understand how the Earth system transitioned into climatic catastrophes in the Cryogenian (Figure 1).
Geological timescale showing the Tonian Period (yellow) between the ‘boring billion' (pink) and ‘freezing millions' (blue).
Emerging geochemical data indicate that atmospheric pO2 levels and oceanic redox structures may have experienced major changes in the Tonian Period  (Figure 1). Planavsky et al.  specifically proposed that atmospheric pO2 levels were prohibitively low for animal evolution until around 800 Ma (but see [10,11]). Guilbaud et al.  showed that, while deep ocean waters may have remained anoxic and ferruginous, mid-depth seawaters shifted from predominantly sulfidic to largely ferruginous conditions in the Tonian Period. Reinhard et al.  argued that the late Tonian to Cryogenian Period (800–635 Ma) is characterized by intermediate pO2 levels and a transitional redox state between pervasively ferruginous and largely oxygenated deep oceans. They further hypothesized that fundamental changes in the global phosphorus and nitrogen cycles may have occurred in the late Tonian Period, due to the combined effects of a weaker deep-sea Fe-P trap (e.g. reduced removal of phosphorus from ferruginous deep waters as vivianite or green rust species) and enhanced production of nitrate that feeds denitrification, resulting in nitrogen limitation on biospheric productivity on geological timescales . It is expected that these changes would have impacted biospheric evolution and stimulated evolutionary innovations , which may have in turn influenced further changes in oceanic redox structures . Thus, it is critical to examine the Tonian fossil record and to more accurately document large-scale diversity patterns and evolutionary innovations in the Proterozoic, in order to explore the possible correlations and links between biospheric and environmental evolution during this key geological period.
Molecular clock estimates and biomarker fossils provide additional impetus to investigate the Tonian fossil record. Molecular clock data suggest that crown-group animals may have diverged in the Tonian Period [15,16], raising the intriguing possibility of close coupling between redox evolution and evolutionary innovations. Biomarker fossils indicate that marine eukaryotic algae did not come to ecological dominance until the Cryogenian and Ediacaran periods , although molecular clock and available paleontological data suggest that eukaryotic algae did evolve before the Tonian Period [3,4,18,19]. Thus, there are compelling reasons to bring the Tonian in the forefront of paleobiological investigation in order to better understand the co-evolution of life and the physical environment during this critical geological period. In this contribution, we summarize recent advances in Tonian paleobiology and identify opportunities for future research.
Large-scale diversity patterns of Tonian eukaryotic fossils
Tonian paleontological studies have been focused almost exclusively on microfossils, particularly organic-walled acritarchs, many of which have been interpreted as unicellular eukaryotes. A recent compilation of Proterozoic acritarchs and other eukaryotic fossils  shows that the assemblage-level taxonomic diversity (analogous to alpha diversity, or species richness at a site or locality) of Tonian assemblages, with a few exceptions, is comparable to that of Mesoproterozoic assemblages, but is significantly lower than that of Ediacaran ones (Figure 2A). Cryogenian assemblages tend to have lower taxonomic diversity, but this may be in part related to the dominance of coarse-grained and poorly fossiliferous glacial deposits in this geological period.
Large-scale diversity trends of Proterozoic eukaryotes.
Existing compilations of global taxonomic diversity (analogous to gamma diversity, or total diversity at a global scale) show that the taxonomic diversity of Tonian acritarchs is significantly higher than in the Mesoproterozoic Era but lower than in the Ediacaran Period . However, these compilations were completed more than 20 years ago and does not reflect the many recent discoveries of Proterozoic microfossils. A recent analysis focusing on Neoproterozoic data and using CONOP—an ordination technique to optimize the order of fossil appearances in the rock record—reveals that the rarefied global diversity shows an overall decreasing trend in the second half of the Tonian Period . This decline precedes Cryogenian glaciations and is therefore unlikely to have been driven by glaciations.
Any estimate of taxonomic diversity of microfossils is necessarily susceptible to problems related to taxonomic treatment of fossil taxa (e.g. taxonomic splitting vs. lumping). To alleviate this problem, Huntley et al.  assembled a morphological database of Proterozoic acritarchs and carried out a morphometric analysis. Their results show that, although the documented morphological range of Tonian eukaryotes is greater than that of the Mesoproterozoic Era (Figure 2B), the morphological disparity as measured by dissimilarity and by variance is only marginally greater in the Tonian than in the Mesoproterozoic Period (Figure 2C). In contrast, both the morphological range and disparity are lower in the Tonian than in the Ediacaran Period, echoing the evolutionary pattern of taxonomic diversity.
Proterozoic macrofossils are typically preserved as carbonaceous compressions. They are relatively rare and their interpretation as eukaryotes are sometimes controversial. For example, the coiled macrofossil Grypania from the Paleo-Mesoproterozoic Era has been variously interpreted as a eukaryote  or a giant cyanobacterium . Other phylogenetically unresolved carbonaceous macrofossils include the common Tonian fossils Chuaria and Tawuia [26,27], as well as various forms from the Paleo-Mesoproterozoic successions in North China [28–30]. These problems make it difficult to obtain a reliable picture of large-scale diversity pattern of Proterozoic eukaryotic macrofossils. The only compilation of Proterozoic carbonaceous compression macrofossils was published more than a decade ago . Despite the limited number of macrofossils included in this database, the analysis shows that Tonian macrofossils do occupy a greater morphological range than Mesoproterozoic ones (Figure 2D), although the morphological disparity as measured by variance is comparable between Tonian and Mesoproterozoic, and the morphological range and disparity are both lower in the Tonian than in the Ediacaran Period (Figure 2E). Cryogenian carbonaceous compression macrofossils  are too few to warrant any meaningful analysis.
To summarize, the fossil record shows a consistent pattern that the Ediacaran Period exhibits greater taxonomic diversity (both at assemblage and global levels), morphological range, and morphological disparity than other geological intervals in the Proterozoic Eon. There seems to be a Tonian increase, relative to the Mesoproterozoic Era, in global taxonomic diversity and morphological range, although this increase is not apparent or significant in terms of assemblage-level taxonomic diversity and morphological variance. Finally, it is likely that the Tonian as a geological period may have greater diversity than the Cryogenian, although there may be second-order dynamics or short-term changes within the Tonian Period .
Evolutionary innovations in the Tonian Period
Important Tonian microfossils that have the potential to shed light on evolutionary innovations include the following: (1) various ornamented acritarchs such as Trachyhystrichosphaera (Figure 3A), Culcitulisphaera, and Cerebrosphaera, which are widely accepted as eukaryotic organisms [33–37]; (2) multicellular eukaryotes such as Proterocladus (Figure 3B), which has been interpreted as a siphonous green alga ; Palaeovaucheria (Figure 3C) and Jacutianema, which have been interpreted as putative xanthophytes or members of the stramenophiles ([39,40], but see ); and Cheilofilum, which has been compared with the extant ascomycete fungus Annellophora (, but see ); (3) vase-shaped microfossils or VSMs (Figure 3D) that have been interpreted as members of amoebozoans and rhizarians [43–45]; (4) apatitic scale microfossils or ASMs (Figure 3E) that are interpreted as remains of biomineralizing eukaryotes .
Representative Tonian fossils.
Tonian macrofossils are relatively few, but they are potentially important in elucidating the evolution of multicellularity and cell differentiation. These include the morphologically simple and stratigraphically long-ranging genera Tawuia and Chuaria (Figure 3F–G), which may be polyphyletic and include organisms with a multicellular stage in their life cycle . The Tonian macrofossil Longfengshania (Figure 3H) shows evidence of morphological differentiation into a holdfast, a stipe, and a spherical to ellipsoidal thallus, suggesting an affinity with morphologically differentiated multicellular algae based on overall morphological similarities to modern algal analogs . Of significant interest are certain ribbon-shaped macrofossils with transverse annulations, including Sinosabellidites (Figure 3I), Protoarenicola, Pararenicola (Figure 3J), and Parmia [48–50]. These fossils have been variously interpreted as wormlike animals [50,51] or siphonous algae [49,52].
Two recent discoveries in Tonian paleontology warrant special highlight here because of their ecosystem-wide implications. Porter  reported circular and semicircular holes on acritarch vesicles and VSM tests (Figure 3D,K). These holes may have been made by predatory eukaryotes, suggesting the possible presence of eukaryovory (i.e. eukaryotes preying on eukaryotes). It has been proposed that eukaryovory and other ecological interactions may have driven the evolution of novel biochemistry and escalated eukaryote diversification in the Tonian Period . Pang et al.  reported a filamentous fossil (Anhuithrix magna) with differentiated cells, and interpreted it as a cyanobacterium with differentiated akinetes (or resting cells) and by implication also heterocysts (or cells specialized for nitrogen fixation) (Figure 3L–M). Because akinetes and heterocysts are only found in the cyanobacterial clade of Nostacales (or subsections IV + V), Anhuithrix magna provides a minimum age constraint on the divergence time of the Nostacales. Anhuithrix magna bolsters the Proterozoic fossil record of cyanobacterial akinetes, previously represented by the Paleoproterozoic–Mesoproterozoic genus Archaeoellipsoides [56,57], whose akinete interpretation has been contested . Improving the fossil record of akinetes is important, not only because akinete fossils serve key calibrations for molecular clock estimates of the Nostacales [18,59,60], but also because the development of heterocysts has been hypothesized to be an evolutionary innovation in response to rising pO2 levels (, but see discussion below).
To summarize, four extant eukaryote clades (Opisthokonta, Amoebozoa, Plantae, and SAR) may be represented by Tonian fossils, which are regarded as crown-group eukaryotes (Figure 4). These clades thus may have diverged during or before the Tonian Period. The presence of green algae (along with the late Mesoproterozoic red alga Bangiomorpha ) and putative xanthophytes in the Tonian Period places minimal age constraints on the origin of primary and secondary plastids, which are key innovations in the evolution of photosynthetic eukaryotes. There is also convincing paleontological evidence for the evolution of eukaryote multicellularity, cell differentiation, biomineralization, eukaryovory, and cyanobacterial heterocysts during or before the Tonian Period  (Figure 4).
Simplified phylogenetic tree of eukaryotes and Tonian fossils.
To more accurately characterize the big-picture evolutionary patterns of eukaryotes in the Tonian Period, it is imperative to carry out comprehensive compilation and analysis of both microfossils and macrofossils throughout the entire Proterozoic with the most recent taxonomic treatments and geochronological constraints, to address problems of sampling biases, and to analyze the database with multiple methods and metrics (e.g. both taxonomic and morphological diversity). Some compilations were made more than a decade ago [21,23,31,61], and need to be updated to account for more recent discoveries and new age constraints. Others are more up to date [20,22,62], but they either do not cover the entire Proterozoic Eon or do not contain global diversity data. Among these compilations, only a few recorded morphological characters (as opposed to taxonomic listings) [23,31], yet morphological diversity (independent of taxonomic diversity) is an essential measure of evolutionary success, can provide important insights into large-scale evolutionary patterns, and can alleviate taxonomic problems noted above. With few exceptions [20,22,23], previous analyses have not adequately addressed the problem of sampling biases. Statistical methods have been developed to address sampling biases in the Phanerozoic fossil record, and it is time to carry out comprehensive statistical analysis of the Tonian fossil record in order to adequately address sampling biases.
Phylogenetic interpretation of many Proterozoic acritarchs and carbonaceous macrofossils remains a significant challenge. Only a handful of Tonian fossils have been phylogenetically resolved to major eukaryotic groups. Many are thought to be eukaryotes, but their phylogenetic positions within the total-group eukaryotes are unknown. Some are not resolved even at the broadest phylogenetic level (e.g. bacteria vs. eukaryotes; crown- vs. stem-group eukaryotes, the latter of which were probably common in Proterozoic oceans ). It has been shown recently that detailed taxon-specific microstructural, ultrastructural, δ13Corg, and biomarker analyses of individual fossils can provide important insights into their phylogenetic affinities (e.g. [64–70]. This line of research is currently hampered by the incomplete survey of both Proterozoic fossils and extant analogs, and it presents opportunities to develop novel proxies for phylogenetic interpretation and to test the phylogenetic specificity of existing proxies.
In light of recent molecular clock estimates of a Tonian divergence of crown-group animals [15,16], it is perplexing that thus far no convincing animal fossils have been identified in the Tonian Period. On the other hand, it can be difficult to conclusively identify early animal fossils, because they may have been soft-bodied organisms and unique animal apomorphies have little likelihood of being preserved in the fossil record. Cohen et al.  proposed that ultrastructural features of acritarchs can provide phylogenetic insights and they interpreted certain Ediacaran spinose acritarchs as animal resting cysts on the basis of ultrastructures. Few Tonian spinose acritarchs have been investigated for ultrastructures. In this regard, Tonian spinose acritarchs such as Trachyhystrichosphaera (Figure 3A) and ‘Tappania'  present excellent opportunities for detailed ultrastructural, δ13Corg, and biomarker analyses in order to constrain their phylogenetic affinities. The same can be said of Tonian carbonaceous compression macrofossils such as Sinosabellidites, Protoarenicola, Pararenicola, and Parmia, which were once interpreted as wormlike animals [50,51] but later as siphonous algae [49,52].
Not only animal and eukaryotic fossils, but also bacterial and particularly cyanobacterial fossils, can provide important insights into biological–environmental co-evolution in the Tonian Period. For example, the recent discovery of akinete-bearing and mat-forming filamentous cyanobacteria from Tonian rocks  may have implications for local or even global redox conditions . These akinete-bearing cyanobacteria may have also borne specialized heterocysts for nitrogen fixation, because modern akinetes occur almost exclusively in heterocystous cyanobacteria  and modern akinete-bearing cyanobacteria are phylogenetically nested within a clade of heterocystous cyanobacteria . The universal enzyme for nitrogen fixation, nitrogenase, is highly sensitive to oxygen and can be irreversibly inactivated in the presence of free oxygen [74,75]. The degree of sensitivity is highly variable, with some diazotrophs showing greater tolerance to oxygen than others . This variability is partly due to the development of various physiological and biochemical strategies to protect nitrogenase . One of these strategies is the development of thick-walled nitrogen-fixing heterocysts in diazotrophic cyanobacteria , which have to cope with the conflicting functions of O2-producing photosynthesis and O2-sensitive nitrogen fixation. Among modern diazotrophic cyanobacteria, the optimal ambient O2 concentration for nitrogen fixation can vary from <12 µM in non-heterocystous forms to 300 µM in heterocystous forms , but nitrogenase activity drops below 10% of optimal levels when ambient dissolved O2 levels are above 600 µM .
As most cyanobacteria carry out oxygenic photosynthesis, it is expected that oxygen concentrations within cyanobacterial cells and mats are higher than the ambient environment. This can be assessed by estimating the excess O2 concentration, or the within-cell or within-mat O2 concentration that is above the ambient O2 level. For planktonic unicellular cyanobacteria, the excess intracellular oxygen due to photosynthesis is related to the photosynthesis rate (which controls O2 production) and cell size (which controls diffusion), but it is negligible (<0.25 µM for cells <1 µm in diameter) due to effective diffusion of oxygen to the ambient environment . However, because of enhanced O2 production and impeded diffusion, excess oxygen concentrations in cyanobacterial colonies and benthic mats can reach 500 µM or higher during daytime [78–80]. Indeed, O2 bubbles can form in modern cyanobacterial mats , and possible O2 bubbles have been identified in Proterozoic cyanobacterial mats . Thus, the main challenge for diazotrophic cyanobacteria that form colonies, aggregates, and microbial mats is the local production and diffusion of O2, rather than the overall atmospheric pO2 levels or ambient dissolved O2 concentrations per se. In other words, even if atmospheric pO2 levels were low (e.g. in the Paleo- and Mesoproterozoic), there may still be an environmental pressure for the development of physiological and biochemical strategies to protect nitrogenase as long as cyanobacteria form colonies, aggregates, or microbial mats. That said, it remains possible that heterocysts in diazotrophic cyanobacteria may have been an evolutionary innovation in response to the environmental ramifications of rising pO2 levels [9,13]. The challenge for paleontologists is to determine whether mat-forming heterocystous cyanobacteria began to diversify in the Tonian Period or much earlier [55,57,58].
To summarize, a key question driving future research of Tonian paleontology is whether the Tonian rise in pO2 levels left a preservable and identifiable paleontological signature. This can be in the form of diversity patterns (e.g. significant Tonian increase in taxonomic and morphological diversity) or evolutionary innovations (e.g. rise of animals, appearance of certain redox-sensitive physiologies). Addressing this question will not only lead to a more comprehensive picture of the Tonian Earth system, but will also inform us about the evolutionary pace in the wake of the ‘boring billion' and set a baseline for studying the biological impact of the ‘freezing millions' that ensued.
The Tonian Period is critical to understanding the transition from the Mesoproterozoic ‘boring billion' to the Cryogenian ‘freezing millions'.
There seems to be a Tonian increase in global taxonomic diversity and morphological range relative to the Mesoproterozoic Era.
Crown-group eukaryotes and major eukaryote groups (including Opisthokonta, Amoebozoa, Plantae, and SAR) may have already diverged by the Tonian Period.
Important evolutionary innovations (possible eukaryovory, eukaryote biomineralization, and cyanobacterial heterocysts) may have arisen by the Tonian Period.
Thus far, no convincing animal fossils have been found in the Tonian Period.
The authors would like to thank National Science Foundation Sedimentary Geology and Paleobiology Program [EAR1528553] and NASA Exobiology and Evolutionary Biology Program [NNX15AL27G] for providing financial support; guest editors Tim Lyons, Mary Droser, and Susannah Porter for inviting us to contribute; and two anonymous reviewers for constructive comments.
The Authors declare that there are no competing interests associated with the manuscript.