News and Views

Fossil eukaryotes: Fungal origins?

Newly discovered filamentous fossils from 2.4-billion-year-old oceanic lavas suggest that eukaryotes of possible fungal affinity are much older than previously thought.

Palaeontological evidence reported by Bengtson and colleagues1 from Palaeoproterozoic sub-seafloor lavas of South Africa challenges current thinking about when and where eukaryotes evolved (Fig. 1). Intriguing filamentous microfossils are reported from mineralized vesicles in the lavas that form branching networks and are interpreted to be fungal-like, raising the possibility of an early sub-seafloor origin for fungi. Independent of whether these fossils are definitely fungal or not, these findings pose provoking questions about the antiquity of eukaryotes during an interval of time known as the Great Oxidation Event of Earth's atmosphere approximately 2.4 to 2.2 billion years ago (Ga)2.

Figure 1: Geological timeline.
Figure 1

The occurrence of newly discovered fungus-like fossil filaments in sub-seafloor lavas that are coincident with the onset of Earth's atmospheric oxygenation approximately 2.4 Ga (ref. 1). The Ongeluk filaments described by Bengtson et al. are a billion years older than fossil eukaryotes found in sedimentary rocks at 1.4 Ga (ref. 10), and several times older than filamentous fossil networks found in Phanerozoic lava vesicles at approximately 385 Ma (ref. 5). Also shown are organic microfossils interpreted as probable fungi from sedimentary rocks dated at 850 Ma (ref. 6). Images (left to right) reproduced with permission from ref. 1, Macmillan Publishers Ltd; ref. 10, Cold Spring Harbor Lab Press; ref. 6, Cambridge Univ. Press; ref. 5, Wiley.

In recent years, scientists have made significant advances in exploration of the deep sub-seafloor uncovering a vast microbial biosphere. We now appreciate that fluid–rock interaction in the oceanic crust supports diverse microbial communities that include heterotrophic organisms such as fungi3. A combination of microbiological data, textural and geochemical evidence demonstrates a thriving sub-seafloor biosphere. Filamentous microtextures found in mineralized cavities and fractures within seafloor lavas, as well as microscopic tunnels found in volcanic glass, have been interpreted as representing the fossilized remains of endolithic microorganisms. In the geological record comparable mineral encrusted filaments from vesicles (former gas bubbles) and fractures in sub-seafloor lavas have been reported from the Eocene (46 Ma) to the Devonian (385 Ma)4,5, suggesting that the oceanic crust represents an ancient habitat for life. The work of Bengtson et al. extends this evidence into much older (2.4 billion years) sub-seafloor lavas and suggests a fungal interpretation. These findings greatly predate current estimates for the earliest putative fossil fungi from sedimentary rocks dated at 1.4 Ga (ref. 6).

In their study, Bengtson et al. provide compelling arguments for a biological and early seafloor origin of the fossilized filaments. Their samples come from more than 800 metres beneath the ancient seafloor in lavas of the Ongeluk Formation in the Transvaal Supergroup of South Africa, and from the start of an important interval recording the first accumulations of significant atmospheric oxygen2. The authors present synchrotron-based imaging of the fossilized filaments to reveal their intricate morphology in three dimensions. The filaments branch and connect to form networks indicative of a high level of complexity supporting a biogenic and eukaryotic origin. State-of-the-art electron microprobe mapping is used to show that the filaments formed in the Palaeoproterozoic sub-seafloor and were preserved by early generations of calcite and chlorite that precipitated in vesicles and fractures in the lavas. The authors address the majority of criteria that have been proposed as necessary to demonstrate the biogenicity of sub-seafloor traces of life7. Briefly, these include: evidence of primary filamentous growth with size, shape and orientations consistent with biological growth and that exclude abiotic mechanisms of tunnel or filament formation. Some palaeontologists will be troubled by the absence of organics from the Ongeluk filaments, which are traditionally taken as a requirement for biogenicity, however, in permeable sub-seafloor environments organic matter is unlikely to be preserved. Bengtson et al. also present strong textural and mineral chemical evidence to confirm that the filaments formed during early seafloor alteration of the lavas, and are unrelated to much younger metamorphic or fluid events (compared with ref. 8). Sceptics may contend that we do not understand the full range of mineral growth and recrystallization processes that can occur in void spaces in sub-seafloor lavas, and that these might conceivably be able to produce abiotic filaments. Such processes seem incapable, however, of explaining the intricate morphologies exhibited by the Ongeluk filaments, or the mineralization sequence, which is consistent with alteration in an early sub-seafloor environment.

The taxonomic affinity of the Ongeluk filaments is not fully resolved. Bengtson et al. highlight features such as anastomosis of the filaments in interlocking networks that are suggestive of fungal hyphae, as well as bulbous protrusions concentrated near the base of the filaments that are reminiscent of fungal spores. They argue, on the basis of these morphological, textural and growth comparisons to mineralized filaments found in much younger Phanerozoic lavas, that the Ongeluk filaments could represent fossilized fungi, and exclude other groups such as Actinobacteria due to their difference in size. The authors do acknowledge, however, the possibility that we may be unaware of extinct or uncultured eukaryotic groups that could grow filaments with similar morphological complexity to fungal mycelia networks. This caveat is shared by younger claims for fossil fungi in the sedimentary rock record6, and so future work is needed to improve our knowledge of the taxonomy of cavity-dwelling mycelial-forming organisms. Nonetheless, the implications of finding potential fungi 2.4 billion years ago remain significant, not least because the early fossil record of fungi is sparse, and the molecular clock estimates for the divergence of fungi highly variable9.

The work of Bengtson et al. raises the question of whether we have been looking in the wrong place for the earliest eukaryotes and fossil fungi in particular. Perhaps we have. Fungi are widely thought to have originated in aquatic, non-marine environments and the evidence for Precambrian terrestrial life is largely based on geochemical data from ancient soil profiles known as palaeosols, where robust fossil evidence for microbial and fungal activity is lacking. In contrast, the Proterozoic deep sub-seafloor would appear to offer a more sheltered and conserved environment for the emergence of life that has now been shown to preserve textural evidence of a deep biosphere. Future investigations of Proterozoic and late Archaean lavas will help to explore if indeed fungi originated in the sub-seafloor.

The study of Bengtson and colleagues offers new perspectives on the early fossil record of eukaryotes as a whole (Fig. 1). The Proterozoic includes well-accepted fossilized red algae, crown group eukaryotes dated at 1.2 Ga, as well as older stem group eukaryotes from between 1.6 and 1.8 Ga (ref. 10), and controversial macrofossils of debated eukaryotic affinity from rocks that are 2.1 billion years old11. Future discoveries of sub-seafloor filamentous fossils may offer a new approach to resolving the early fossil record of eukaryotes, raising fundamental questions regarding how large size and relative complexity could be achieved during the early stages of Earth's atmospheric oxygenation.


  1. 1.

    et al. Nat. Ecol. Evol. 1, 0141 (2017).

  2. 2.

    et al. Proc. Natl Acad. Sci. USA 114, 1811–1816 (2017).

  3. 3.

    & Fungal Ecol. 5, 463–471 (2012).

  4. 4.

    , , & Geomicrobiol. J. 21, 241–246 (2004).

  5. 5.

    , , & Geobiology 6, 125–135 (2008).

  6. 6.

    Paleobiology 31, 165–182 (2005).

  7. 7.

    , , , & Geobiology 8, 245–255 (2008).

  8. 8.

    & Proc. Natl Acad. Sci. USA 111, 8380–8385 (2014).

  9. 9.

    , , & BMC Evol. Biol. 4, 2 (2004).

  10. 10.

    Cold Spring Harb. Persp. Biol. 6, a016121 (2014).

  11. 11.

    et al. Nature 466, 100–104 (2010).

Download references

Author information


  1. Nicola McLoughlin is in the Department of Geology and Albany Museum, Rhodes University, Grahamstown, 6140, South Africa.

    • Nicola McLoughlin


  1. Search for Nicola McLoughlin in:

Competing interests

The author declares no competing financial interests.

Corresponding author

Correspondence to Nicola McLoughlin.