The earliest strong evidence for photosynthesis on Earth dates to roughly 3.5 billion years ago, based on fossilized microbial structures found in ancient rocks from Western Australia and South Africa. Hints from carbon isotope signatures push the possible origin even further back, to around 3.8 billion years ago, though those oldest signals remain contested. What makes the question so tricky is that photosynthesis did not arrive as a single invention: it evolved in stages, with simpler forms appearing long before the oxygen-producing version that transformed the planet.
What the Oldest Rocks Tell Us
The search for ancient photosynthesis relies on indirect clues preserved in rocks that have survived billions of years of geological recycling. The two main lines of evidence are fossilized microbial mats called stromatolites and the chemical fingerprint that living organisms leave in carbon.
Stromatolites are layered, dome-shaped structures built by communities of microbes, often photosynthetic ones, that trap sediment and precipitate minerals. The oldest convincing examples come from rocks in Western Australia and South Africa that formed around 3.5 billion years ago. Based on comparisons with modern stromatolites and their microbial communities, these structures suggest that light-harvesting organisms were already active by that time.1PubMed. The oldest records of photosynthesis
Carbon isotopes offer a separate and potentially older line of evidence. When organisms fix carbon dioxide into organic matter, they preferentially take up the lighter carbon-12 over the heavier carbon-13. This leaves organic carbon with a distinctive isotopic signature. In the Isua rocks of southwestern Greenland, some of the oldest sedimentary rocks on Earth at roughly 3.8 billion years old, researchers have found reduced carbon with isotopic values in the range characteristic of biological carbon fixation.2Precambrian Research. Carbon isotopes as biogeochemical recorders of life over 3.8 Ga of Earth history: evolution of a concept Metamorphosed shales from West Greenland of a similar age contain carbon with isotopic values consistent with a planktonic biological source.3Earth and Planetary Science Letters. U-rich Archaean sea-floor sediments from Greenland – indications of >3700 Ma oxygenic photosynthesis
There is a serious catch, though. The Isua rocks have been cooked and squeezed by metamorphism for billions of years, and that processing makes it harder to trust the isotopic signal at face value. Several non-biological chemical processes can produce carbon with the same kind of isotopic depletion, so the mere presence of light carbon is not enough to confirm that life, let alone photosynthesis, was responsible.4ScienceDirect (Elsevier). The Archean Earth – Chapter 5.3 – Origins of Archean organic matter This is one of the most contentious debates in early-Earth science: the 3.8-billion-year claim rests on geochemistry that can be read more than one way.
Microfossils and the Problem of Lookalikes
You might expect that ancient fossils of photosynthetic cells would settle the question. In practice, identifying genuine microfossils in rocks billions of years old is extraordinarily difficult. The most famous example is the Apex chert from Western Australia, roughly 3.46 billion years old, which was initially reported to contain filamentous microfossils resembling cyanobacteria. That interpretation has since been challenged. Detailed comparisons with well-preserved microfossils from younger rocks showed that the Apex filaments lack the cylindrical cell structure, consistent diameter, and carbon distribution expected of genuine microbial fossils. Instead, they may be mineral artifacts, possibly exfoliated mineral grains coated in carbon.5Journal of the Geological Society. Comparative multi-scale analysis of filamentous microfossils from the c. 850 Ma Bitter Springs Group and filaments from the c. 3460 Ma Apex chert
The lesson is that rocks this old are full of structures and chemical signatures that can look biological but turn out not to be. Researchers have learned to apply increasingly strict criteria before accepting something as a genuine trace of life. That skepticism is why the 3.5-billion-year stromatolite evidence is considered more robust than the 3.8-billion-year carbon isotope evidence: stromatolites combine physical structure with chemistry, making it harder for a purely geological process to fake.
The First Photosynthesis Did Not Make Oxygen
When most people think of photosynthesis, they picture plants splitting water and releasing oxygen. But the earliest forms of photosynthesis almost certainly did not use water at all. Instead, they used other electron donors, particularly hydrogen sulfide and dissolved iron, to drive the process. This “anoxygenic” photosynthesis does not produce oxygen as a byproduct, and it is still practiced today by several groups of bacteria.
One influential proposal is that the first light-driven metabolisms used hydrogen sulfide as an electron source, powered by a single type of reaction center. Over time, organisms developed two linked reaction centers, eventually acquiring the ability to split water. In this view, sulfur-based photosynthesis served as a bridge between the earliest chemically powered life and the later evolution of water-splitting, oxygen-producing photosynthesis.6FEMS Microbiology Reviews. A physiological perspective on the origin and evolution of photosynthesis
Iron-oxidizing anoxygenic photosynthesis, sometimes called photoferrotrophy, is particularly relevant to the geological record. During the Archean, Earth’s oceans were rich in dissolved iron and dissolved silica but contained essentially no free oxygen. Bacteria that used light energy to oxidize dissolved iron would have produced iron minerals as a waste product. This process is one leading explanation for the massive banded iron formations found in rocks dating from about 3.8 to 1.8 billion years ago. Laboratory experiments have shown that in silica-rich water resembling ancient seawater, photoferrotrophic bacteria produce iron oxides that separate from the cells themselves, leading to iron-rich, organic-poor sediment layers that match the chemistry of banded iron formations remarkably well.7PubMed Central. Photoferrotrophy, deposition of banded iron formations, and methane production in Archean oceans
Modern analogues support this picture. In an iron-rich, sulfide-poor lake studied as an Archean ocean stand-in, green sulfur bacteria sustain themselves by oxidizing dissolved iron in the presence of light, exactly as ancient photoferrotrophs are proposed to have done. The lake’s chemistry and microbial community suggest that similar populations once lined the sunlit layers of ancient oceans, driving iron deposition and fueling marine productivity.8PubMed Central. Photoferrotrophs thrive in an Archean Ocean analogue Fossilized remains of iron-oxidizing bacteria have even been identified within banded iron formations themselves, adding direct physical evidence to the geochemical argument.9Nature Communications. Fossilized iron bacteria reveal a pathway to the biological origin of banded iron formation
Which Light-Harvesting Pigments Came First
A related puzzle involves chlorophyll and bacteriochlorophyll, the pigments that capture light energy. The standard expectation used to be that chlorophyll, the pigment used by plants and cyanobacteria for oxygenic photosynthesis, came first on the biosynthetic pathway and therefore evolved first. Bacteriochlorophyll, used by anoxygenic photosynthesizers, requires additional chemical reduction steps, so it seemed like a later, more derived product.
Molecular evidence flipped that assumption. Analyses of the genes encoding the enzymes responsible for pigment biosynthesis indicate that bacteriochlorophyll likely evolved before chlorophyll.10PubMed. Molecular evidence for the early evolution of photosynthesis This finding contradicts the older Granick hypothesis and aligns with the broader picture that anoxygenic photosynthesis predated the oxygenic kind.11PubMed. Evolution of chlorophyll and bacteriochlorophyll: the problem of invariant sites in sequence analysis Essentially, the pigments that absorb longer, lower-energy wavelengths of light and work without producing oxygen appear to be the more ancient versions.
When Water Splitting Began
The jump to oxygenic photosynthesis, the kind that splits water and releases oxygen, was one of the most consequential innovations in the history of life. It required the evolution of a remarkable molecular machine: a cluster of four manganese atoms and one calcium atom that catalyzes the extraction of electrons from water. This catalytic center, embedded in what is now called Photosystem II, appeared roughly three billion years ago and has been called the “big bang of evolution” because of the cascade of changes it set off.12Biochemistry. Mn4Ca Cluster of Photosynthetic Oxygen-Evolving Center: Structure, Function and Evolution
Interestingly, some researchers have pointed out that the structural foundations for water oxidation may have been present much earlier than the actual capability appeared. The simplest known reaction center, found in modern heliobacteria (an anoxygenic group considered to retain ancient features), already contains a calcium-binding site that structurally resembles the manganese-calcium cluster of Photosystem II. This suggests that the molecular scaffolding for water splitting was present in very early reaction centers, and that relatively few modifications were needed to make the leap to oxygen production.13Trends in Plant Science. When Did the First Photosynthesis Emerge on Earth?
Whiffs of Oxygen Before the Atmosphere Changed
Even after oxygenic photosynthesis evolved, oxygen did not accumulate in the atmosphere for a long time. For hundreds of millions of years, oxygen was produced locally but consumed almost immediately by reactions with iron, sulfur, and organic matter. The permanent rise of atmospheric oxygen, known as the Great Oxidation Event, did not occur until around 2.4 to 2.3 billion years ago.
One of the strongest indicators of the pre-oxygenation atmosphere comes from sulfur chemistry. Rocks older than about 2.3 billion years contain sulfur isotope patterns called mass-independent fractionation, which can only form in an atmosphere with essentially no free oxygen. The disappearance of this signal in younger rocks marks the point when atmospheric oxygen crossed a threshold, though the exact cause of the transition is debated: it could reflect the rise of oxygen itself or a collapse in atmospheric methane that had previously helped maintain anoxic conditions.14PubMed. Mass-independent fractionation of sulfur isotopes in Archean sediments: strong evidence for an anoxic Archean atmosphere15Geobiology. The loss of mass‐independent fractionation in sulfur due to a Palaeoproterozoic collapse of atmospheric methane
But traces of oxygen do show up before the main event. In the Mount McRae Shale of Western Australia, sediments dating to about 2,501 million years ago contain enrichments of molybdenum and rhenium, metals that are mobilized by oxidative weathering. Their presence in marine sediments this old points to small amounts of free oxygen in the environment more than 50 million years before the Great Oxidation Event formally began.16PubMed. A whiff of oxygen before the great oxidation event? Broader analyses of the Archean sedimentary record support the idea that oxygen levels were vanishingly low but still well above what you would expect on a planet with no biological oxygen production at all.17PubMed Central. Reconciling evidence of oxidative weathering and atmospheric anoxia on Archean Earth
The gap between the likely origin of oxygenic photosynthesis, somewhere around 3 billion years ago, and the Great Oxidation Event at 2.4 billion years ago is one of the most puzzling features of Earth history. It means the planet had organisms producing oxygen for hundreds of millions of years before that oxygen permanently changed the atmosphere. The likely explanation is that geological oxygen sinks, particularly volcanic gases and dissolved iron in the oceans, consumed oxygen as fast as it was produced, until those sinks were gradually exhausted.
What Molecular Clocks Say, and Why They Disagree
Fossils and geochemistry tell us when certain products of photosynthesis appear in the rock record, but they do not directly date the organisms responsible. Molecular clock analyses try to fill that gap by using the rate of genetic change to estimate when lineages diverged. For cyanobacteria, the group responsible for all oxygenic photosynthesis on Earth, these clocks have produced strikingly different answers depending on the assumptions fed into them.
One analysis concluded that “crown group” cyanobacteria, meaning the common ancestor of all living cyanobacterial species, evolved around 2.0 billion years ago, well after the rise of atmospheric oxygen.18PubMed. Crown group Oxyphotobacteria postdate the rise of oxygen If that date is correct, it would mean the earliest oxygen-producing photosynthesizers belonged to lineages that left no modern descendants, a “stem group” that went extinct after giving rise to the cyanobacteria we know today.
A more recent analysis reached a different conclusion, showing that multiple interpretations of the cyanobacterial fossil record are consistent with an Archean origin of crown-group cyanobacteria, potentially placing their origin before the Great Oxidation Event. By incorporating information from horizontal gene transfers between species as additional dating constraints, this study improved the precision of its age estimates and found support for the older timeline.19PubMed Central. The Archean origin of oxygenic photosynthesis and extant cyanobacterial lineages
The disagreement is not trivial. It reflects genuine uncertainty about how to calibrate molecular clocks for organisms that diverged billions of years ago, with sparse and ambiguous fossil calibration points. Both camps agree that some form of oxygenic photosynthesis existed before the Great Oxidation Event, but whether modern cyanobacterial lineages were already diversifying by then, or whether an earlier, now-vanished group was doing the heavy lifting, remains an open question.
How Photosynthesis Genes Travel Between Species
One reason the evolutionary history of photosynthesis is so tangled is that photosynthesis genes do not always follow the family tree. Bacteria can swap large blocks of genes between unrelated species through horizontal gene transfer. The genes for photosynthesis in certain purple bacteria, for example, were acquired from a distantly related bacterial lineage rather than inherited from a shared ancestor.20PubMed. Horizontal transfer of the photosynthesis gene cluster and operon rearrangement in purple bacteria More broadly, the highly conserved clusters of photosynthesis genes found in aerobic anoxygenic phototrophs show evidence of transfer between species, genera, and even entirely different phyla.21PubMed Central. Characteristics and Evolutionary Analysis of Photosynthetic Gene Clusters on Extrachromosomal Replicons: from Streamlined Plasmids to Chromids
This gene swapping means that the distribution of photosynthesis across the tree of life does not neatly trace a single origin and branching descent. It complicates molecular clock analyses, since the gene tree and the species tree can tell different stories. But it also means photosynthesis had a kind of evolutionary resilience: even if a photosynthetic lineage went extinct, its genetic toolkit could survive in other organisms that had picked it up.
Surviving the UV Onslaught
Early photosynthetic organisms faced a serious environmental challenge that no modern plant deals with. Before oxygen accumulated in the atmosphere, there was no ozone layer. Ultraviolet radiation, including the most damaging short-wavelength UVC rays, poured down on any organism living in shallow, sunlit water, exactly where photosynthesizers needed to be.
Modern cyanobacteria produce a pigment called scytonemin in their outer sheaths that strongly absorbs UVC radiation. Laboratory experiments show that cyanobacteria with scytonemin suffer less photosynthetic damage under UVC exposure, suggesting the pigment likely evolved as an early adaptation to the harsh UV environment of the Precambrian. Its presence would have allowed photosynthetic organisms to colonize shallow-water and possibly even land habitats long before the ozone layer provided protection.22Journal of Phycology. SCYTONEMIN, A CYANOBACTERIAL SHEATH PIGMENT, PROTECTS AGAINST UVC RADIATION: IMPLICATIONS FOR EARLY PHOTOSYNTHETIC LIFE
The Nitrogen Problem
Oxygen was revolutionary, but it also created new biological crises. One of the most important involved nitrogen fixation, the process by which certain microbes convert atmospheric nitrogen gas into a usable form. The enzyme responsible, nitrogenase, is irreversibly damaged by oxygen. So the very organisms that evolved to produce oxygen had to simultaneously figure out how to protect their nitrogen-fixing machinery from the toxic byproduct of their own metabolism. This evolutionary pressure drove the development of spatial and temporal separation strategies in cyanobacteria, such as confining nitrogen fixation to specialized cells or performing it only at night.23PubMed. Geobiological feedbacks, oxygen, and the evolution of nitrogenase
This is a good example of how the origin of photosynthesis was not just a biological milestone but a geobiological one: it changed ocean chemistry, atmospheric composition, and the selective pressures acting on every other organism, including the photosynthesizers themselves.
From Cyanobacteria to Algae and Plants
For roughly two billion years after its origin, oxygenic photosynthesis was exclusively a bacterial affair. The transfer to eukaryotic cells, the lineage that eventually includes algae and plants, happened through endosymbiosis: a non-photosynthetic cell engulfed a cyanobacterium, and instead of digesting it, the two formed a permanent partnership. The cyanobacterium became the chloroplast. Molecular clock analyses place this primary endosymbiosis event sometime before about 1.56 billion years ago, after the split between the lineage leading to plants and algae and the lineage leading to animals and fungi.24Molecular Biology and Evolution. A Molecular Timeline for the Origin of Photosynthetic Eukaryotes
That means photosynthesis spent roughly half of Earth’s history as a purely prokaryotic phenomenon. The familiar green world of leaves, trees, and grasses is a comparatively recent chapter.
Searching for Photosynthesis on Other Worlds
The history of photosynthesis on Earth has become central to astrobiology, particularly the search for life on exoplanets. One of the more promising biosignatures is the “vegetation red edge,” a sharp increase in reflectance at red and near-infrared wavelengths caused by chlorophyll-containing organisms. This spectral feature is usually associated with land plants, which have only been widespread for about half a billion years. But chlorophyll is present in cyanobacteria, algae, lichens, and corals, not just leafy vegetation. Researchers have shown that ocean-surface cyanobacteria and algae could have produced a detectable red edge on Earth more than two billion years ago, and lichens could have extended that signal to at least 1.2 billion years ago.25The Astrophysical Journal Letters. Expanding the Timeline for Earth’s Photosynthetic Red Edge Biosignature
This matters for telescope missions designed to characterize exoplanet atmospheres and surfaces. If a photosynthetic red edge only appeared when land plants did, the window for detecting it on an Earth-like planet would be narrow. But if microbial photosynthesizers can produce the signal, then a planet with life at the cyanobacterial stage, which on Earth lasted for billions of years, might still be identifiable from a distance. The long persistence of photosynthesis on Earth, in one form or another for over three billion years, makes it one of the most enduring and potentially detectable signatures that life can leave on a planet.