What greater magic can there be than the signature trick of all leaf magicians when they extract a rare gas from the atmosphere and turn it into solid wood? Before the industrial revolution, leaves achieved this feat with a concentration of carbon dioxide that was only 0.02 percent, half the level it is today. Every skilled magician weaves a story into their performance, and though devoid of intentionality, plants are no different when they perform the signature trick of photosynthesis.
The narrative plays out an evolutionary history that is written in genetic code, but the script is not inscribed in the cell nucleus with the plant’s chromosomes, where you might expect to find it. The history of photosynthesis is hidden in the tiny bodies called chloroplasts, which are found in their hundreds floating inside every green plant cell. It is chloroplasts that contain the pigment chlorophyll, which make leaves green, and it is inside chloroplasts that the magic of photosynthesis happens.
Leaves are showcases for chloroplasts, designed by natural selection to display them to light, to supply them with the raw materials that they need, and then to rob them of the sugar that they make. I use the term rob deliberately because chloroplasts are captives of plants. The clearest evidence of captivity is that chloroplasts have their own DNA, which they use to multiply themselves, though they cannot do so independently of a plant cell because many of the genes essential to the process were long ago moved to the plant nucleus. How did chloroplasts, which hold all the cards to carbon capture, themselves become captives?
Like the forensic solution to many a murder, the unraveling of this evolutionary mystery is to be found in DNA. The DNA that remains in the chloroplast reveals kinship with a group of photosynthetic microbes called cyanobacteria, many species of which live independent lives in soil or in water. According to the narrative told by chloroplast genes, the relationship between the algal ancestors of land plants and a cyanobacterium began about 1.5 billion years ago. Investigating exactly what happened is tricky, but enough genetic evidence has been preserved to create an Identi-Kit for what the two suspects must have looked like before they joined forces.
Though devoid of intentionality, plants are no different when they perform the signature trick of photosynthesis.
The algal partner, yet to become photosynthetic, must have been a cell that fed on bacteria, perhaps the way that amoebas engulf and devour bacteria today. Algal cells nowadays have a rigid cell wall that prevents them from swallowing anything, but a primitive species called Cymbomonas tetramitiformis has been discovered that fits what we are looking for. This alga goes through a stage in its development when it lacks a rigid cell wall, and it swallows bacteria. The closest genetic fit for the chloroplast ancestor, the swallowee to the swallower, is a cyanobacterium called Gloeomargarita lithophora, which lives glued to rocks in brackish and fresh water. It is easy to imagine how such a photosynthetic bacterium could have been grazed off the rocks by the algal ancestor. Perhaps the first meals were digested, but algae that had poor digestion discovered that the cyanobacteria were more nutritious if allowed to carry on with photosynthesis. Fortunately, we don’t have to prove this hypothesis in a court of law, but scenarios such as I have described are known today in sea slugs. These mollusks suck the chloroplasts out of green algae and sequester them undigested for months inside their cells as a source of photosynthetic energy.
The relationship in which an ancestral alga and a cyanobacterium teamed up is known as symbiosis—meaning “living together.” Though the symbiotic nature of chloroplasts was first suspected in the nineteenth century, not until the late 1960s did it come to be generally accepted. In the present century, now that we have the tools to read gene sequences almost as easily as you are reading the words on this page, another surprise has come to light. The genes for oxygen-producing photosynthesis, unique to cyanobacteria and their chloroplast descendants, turn out to be older than cyanobacteria, and yet the first cyanobacteria were apparently not photosynthetic. What can that possibly mean? By way of an answer, I have a favorite analogy. No one in my family has ever owned a hat, but I’m wearing a really old vintage one. How did I get this hat?
The hat is older than me, so I cannot have made it, and I didn’t inherit it, so I must have acquired it via Horizontal Hat Transfer. Cyanobacteria did not evolve the genes for photosynthesis de novo, they acquired them by horizontal gene transfer. This is the acquisition of genes by some process other than inheritance from parents. It is extremely common among bacteria, and, for example, it is responsible for the rapid spread of genes for antibiotic resistance among bacteria in hospitals. We don’t know where cyanobacteria got the photosynthesis genes from, but they are ancient, stretching back more than 3 billion years. Some of the genes involved have been found carried by viruses that infect bacteria, and these might have been the vector that gave cyanobacteria their unique talent.
Let us pause for a moment beneath the restful green shade of a broad-leaved tree and contemplate the story so far. Over our heads is a palimpsest of evolutionary history containing millions of leaves teeming with billions of chloroplasts. These minuscule green machines are ancient vessels equipped with an even older molecular mechanism that uses the power of sunlight to turn COA and water into sugar and oxygen. The mechanism has been operating for more than three-quarters of the history of life on Earth.
Plants today represent 82 percent of the biomass of life on the planet and sustain all of it. In total, chloroplasts and cyanobacteria fix 100 billion tons of carbon from the atmosphere every year. A little sit-down may be required to take all this in. It’s big, it’s old, but there is still a missing piece of the story. How did we get land plants with leaves? The genes that spin the yarn of evolutionary history like a silk moth spooling out gossamer are endlessly loquacious. From them we learn that oxygen-generating photosynthesis and chloroplasts evolved only once. Through horizontal transfer, chloroplasts are found in some guise in a wide range of marine organisms, from microscopic phytoplankton to giant seaweeds. One of these organisms, a green alga, is the ancestor of all land plants. It belonged to a group called the streptophytes, which had evolved the ability to live in fresh water, an intermediate step between ocean and dry land. But land plants did not stride confidently onto an empty beach like Ursula Andress emerging from the Caribbean Sea in the first Bond film. Oh, No.
Streptophytes were not even the first photosynthetic organisms on land—that prize must go to cyanobacteria, which had been in terrestrial residence for billions of years by the time ancestors of land plants arrived about 500 million years ago. Fossil soils have been found that were formed by microbial mats made of bacteria and fungi, sustained by cyanobacteria. Until Devonian times, 419 million years ago, terrestrial life inhabited a patchwork of drier areas encrusted with microbes such as we find in deserts today, and freshwater and brackish pools inhabited by algae and small marine creatures. Over the late Silurian and early Devonian landscape of encrusted rocks and freshwater wallows towered an extraordinary organism, so huge that its first discoverer believed it to be a tree, which he named Prototaxites (early conifer).
The presence of a tree-size fungus in a world before trees had evolved tells us something important about the terrestrial ecosystem in which it lived.
Prototaxites had a trunk that grew more than eight meters tall and one and a half meters in width, so it’s hardly surprising that it was at first mistaken for a tree. Closer examination ruled this out. And paleobotanists wrangled over whether it was an alga of some kind—green, red, and brown all had their cheerleaders—or maybe a massive lichen. Recent research suggests that Prototaxites was a humongous fungus. Analysis of isotopes in the fossils shows that it did not acquire its carbon directly through photosynthesis, such as an alga or a land plant would have done, but rather it lived on decaying matter and cyanobacterial crusts.
Aside from the intrinsic interest of such an unusual fossil, the presence of a tree-size fungus in a world before trees had evolved tells us something important about the terrestrial ecosystem in which it lived: plenty of carbon was being fixed and turned into organic detritus for these giant fungi to feed upon. The world in which land plants made their debut must therefore have had primitive soils enriched by organic matter, water, warmth, and lots of atmospheric COA. Streptophytes would have been quite at home in the freshwater pools in this landscape, and it has been speculated that these algae may even have become land-adapted themselves. A streptophyte alga called Chlorokybus atmophyticus lives on land among mosses, stones, and in soil, providing a modern example. However it happened, one such streptophyte was the ancestor of both major groups of land plants: the bryophytes (mosses and relatives) and the vascular plants (ferns, conifers, and flowering plants).
Venturing from such tentative beginnings in mud, land plants diversified. We have a beautifully preserved snapshot of life on land 407 million years ago captured in a rock called the Rhynie Chert, named after the tiny Scottish village of Rhynie, not far from the city of Aberdeen, where the rock formation was found. The fossil plants and animals in the chert lived on the margins of a hot-water spring where their bodies were preserved in death by silica. Fossilization in silica produced immortal three-dimensional models that are perfect down to the finest microscopic detail of cellular structure. From these fossils, we know that land plants at this time had diversified into many kinds with no-bigger-than-knee-high, photosynthetic, cylindrical stems, and no leaves. Seven different plant species are fossilized in Rhynie, where there are also traces of Prototaxites, the humongous fungus, cyanobacteria, and aquatic algae.
Living in a modern world canopied by leaves, it’s hard for us to imagine plants without any at all, but leaves were mysteriously late to evolve. The first leaves do not appear in the fossil record until the start of the Devonian period, at least 419.2 million years after the origin of land plants. By the close of the Devonian, 358 million years ago, leaves had evolved independently in at least six different groups of plants. You wait 40 million years for a leafy plant to come along, and then six whole fleets of them arrive at once! What could the explanation be?
Was it mechanically difficult to make flat leaves cantilevered on stalks from existing structures that were all tubular? This seems unlikely because the genes involved in leaf development are ancient and were present in the earliest land plants, even though they were leafless. The comedian Eric Morecambe unwittingly revealed why having all the right genes for leaf development might not produce a leaf. In a classic sketch on TV’s The Morecambe & Wise Show broadcast in 1971, Eric sat down at the piano to play Grieg’s Piano Concerto, under the baton of the celebrated conductor André Previn. Horrified at what he was hearing, Previn confronted Morecambe: “You’re playing all the wrong notes!” Morecambe defiantly replied, “I’m playing all the right notes, but not necessarily in the right order. I’ll give you that.”
During development, genes need to be activated in the right order and in the right place to produce a leaf. This is a matter of gene regulation. Genes, like orchestral players, require the equivalent of a musical score and an André Previn to get them to play together. Small evolutionary changes in regulatory genes can have big effects on structure, so it is difficult to see why evolution of a simple leaf should take 40 million years. By comparison, the rise and diversification of the mammals took only 10 million years.
If there was no inherent genetic or developmental barrier to leaf evolution, could there have been an environmental reason for the delay? At the time plants arrived on land, the climate was hot, and the atmospheric concentration of COA was high. Both of these factors reduce the advantage of a leaf as compared to a stem as a photosynthetic organ. Leaves heat up when exposed to sunlight, and when the air temperature is high, overheating can damage them. By contrast, the smaller surface area of a cylindrical stem and its perpendicular orientation to the direction of sunlight combine to lower the risk of overheating by the sun. On the downside, stems have a disadvantage in light capture, but this is less of a problem when all plants are leafless because they don’t shade one another as much as leafy plants do. The limited surface area of a cylinder potentially also reduces the efficiency of COA capture, but that doesn’t matter so much either if the atmospheric concentration of the gas is high. COA enters leaves through tiny pores called stomata. Early plants had low densities of stomata, suggesting that cylindrical surface area did not limit carbon capture. There was room to accommodate more stomata if required, but plants didn’t use it.
If COA and climate favored stems over leaves 400 million years ago, what changed to trigger the evolution of leaves 40 million years later? The answer is that the concentration of atmospheric COA fell, probably because it was being sequestered by plants. The fall in COA reduced greenhouse warming and cooled the climate. As these environmental changes progressed, the size of leaves found among fossil plants increased. At the same time, the microscopic detail of fossil leaves shows that the density of stomata also increased greatly. The combination of an increase in leaf area and in the density of stomata in the surface of leaves enabled plants to harvest COA more efficiently, compensating for its falling concentration in the atmosphere.
Stomata have a dual function: they admit COA and they also emit water vapor. The increase in the density of stomata that occurred as leaves evolved had the beneficial side effect of increasing evaporative cooling. In modern leaves, evapotranspiration can lower the temperature of a leaf surface to 1°C (2°F) below that of the air. This cooling effect made it possible to have larger leaves, which would have overheated in the high-COA world of earlier times. Once leaf evolution had begun, it had a series of knock-on effects with planet-changing consequences.
_______________________________
Excerpted from Nature’s Magicians: How Leaves Conjure Up Our World by Jonathan Silvertown. Copyright © 2026 by Jonathan Silvertown. Reproduced by permission of Scribner, an imprint of Simon & Schuster. All rights reserved.
Leave a Reply