01. The First Choice

Four billion, two hundred million years ago, give or take — 4.2 Gya, in the shorthand this book will keep using, where Gya means giga-annum, a billion years before now — something on Earth is already alive. The team that reconstructed its chemistry favors a hydrothermal vent or a hot spring as its home: cracked rock, moving water, heat. This population already breathes by fixing carbon dioxide the way an anaerobic acetogen does, an organism that lives by turning carbon dioxide into acetate, through a chemical assembly line called the Wood-Ljungdahl pathway. Every protein its cells make is already built from one hand of amino acid, and only one.

That one-sidedness has a name: chirality, borrowed from the Greek word for hand. An object is chiral when it cannot be laid on top of its own mirror image and matched up, however it is turned. A left hand and a right hand share the same five fingers in the same order, and no rotation ever seats one inside the other's glove. Molecules can be chiral the same way: identical atoms, identical bonds, arranged as two mirror-image versions that no amount of twisting in space will superimpose. Chemists label the two versions L and D. Until 1848, nobody had proven that such mirror-twin molecules were real, physically separable things, rather than a mathematician's trick.

The proof came from a wine cask. Sodium ammonium paratartrate is a crystalline byproduct of winemaking, and by the 1840s chemists already knew it did something strange: dissolved in water, some batches twisted a beam of polarized light as it passed through, and other batches, chemically identical by every test then available, did not. In Paris in 1848, a young chemist named Louis Pasteur sat at a microscope with a pair of tweezers and sorted a heap of the inactive form, one tiny crystal at a time, by the tilt of facets too small to see unaided — some tilted one way, hemihedral to the right, others hemihedral to the left, the mirror opposite.

Then he dissolved the two piles separately and passed each solution through Jean-Baptiste Biot's polarimeter, an instrument built to measure exactly how far polarized light twists on its way through a liquid. In Pasteur's own words, preserved in a 2021 retrospective on the discovery: "I carefully separated the crystal hemihedral to the right and the crystals hemihedral to the left; I observed separately their solutions in Biot's polarizing apparatus, and I saw... the crystals hemihedral to the right deviated the plane of polarization to the right, those hemihedral to the left deviated it to the left." The two piles turned out to be one substance after all — a fifty-fifty mixture of two true mirror-image forms of tartaric acid, sorted apart by hand for the first time in history. Handedness was real. It could be separated, dissolved, and measured.

Life had already made this choice, long before Pasteur found a way to see it. Nearly every protein in nearly every organism alive today is built from L-amino acids, the left-handed form, strung together by the cell's own machinery. Nearly every strand of DNA and RNA rests on a backbone of D-sugars — D-deoxyribose in DNA, D-ribose in RNA — the right-handed form. Nothing in the underlying chemistry obviously demands this split; both hands are equally stable, equally easy to build in a laboratory flask. Life picked one hand for its proteins and the opposite hand for its genetic backbone, and at the level of its core machinery, it has never gone back.

That detail — opposite hands for the two great families of biological molecule — is itself an open question, not a solved one. Any single outside force strong enough to tip chemistry toward one handedness — circularly polarized starlight, a chiral mineral surface acting as a sorting table, even the faint leftward lean built into the weak nuclear force — should be expected to tip everything the same way, not opposite ways for proteins and for sugars. Origin-of-life researchers, writing as recently as 2025 and 2026, still say plainly that there is no widely accepted explanation for why the two major biopolymers ended up on opposite sides of the mirror. It is one of the honest gaps sitting at the center of this whole story.

The rule is strict enough that its exceptions are worth naming one at a time, starting with bacteria. More than ten distinct D-amino acids, chiefly D-alanine and D-glutamate, get built directly into the tough mesh of the bacterial cell wall — peptidoglycan — usually in the sequence L-alanine, D-glutamine, L-lysine (or a related molecule called meso-diaminopimelate), D-alanine, D-alanine. Escherichia coli and Staphylococcus aureus do this today, along with essentially every walled bacterium alive, while archaea and the plant and fungal cells that build walls of their own do not. This reads as a bacterial invention, layered on top of an already-settled handedness, rather than a surviving fragment from before handedness was decided.

A stranger exception belongs to a mammal. Platypus venom carries a dedicated enzyme, roughly 50 to 60 kilodaltons in size, whose entire job is to seize specific amino acids inside the venom's defensin-like peptides and natriuretic peptides — proteins first assembled the ordinary, left-handed way — and flip them to the right-handed form, recognizing a short signature sequence at one end of the chain. Reported in 2006, it was the first such isomerase, an enzyme that converts a molecule into its own mirror-image twin, ever documented in any mammal. The identical trick — build it left-handed, then flip it — evolved again independently in cone-snail venom and in frog skin secretions: three separate lineages, mollusk, amphibian, and monotreme, arriving at the same late workaround long after the original handedness was already fixed.

Exceptions this rare, this late, and this visibly bolted onto standard chemistry make the standard chemistry more striking, not less. Nothing in the physics of an amino acid or a sugar requires the left hand over the right; a coin has no reason to prefer heads. And yet nearly all of life, across billions of years and every branch it has ever grown, kept landing on the same side. A shared, arbitrary choice, preserved this faithfully through so much time and so much divergence, is exactly the kind of signature a family resemblance leaves behind: a clue that one ancestral population handed the choice down, rather than every lineage separately tossing the same coin and happening, again and again, to match.

One Ancestor, or Many Pushed the Same Way

A shared choice is not, on its own, proof of a shared origin. A 2025 review in PNAS lays out the honest counter-argument: an external chiral bias — the same circularly polarized starlight, the same parity-violating weak force, the same kind of chiral mineral surface named above — could in principle have leaned on more than one independent origin-of-life event and tipped every one of them toward the same hand, the way a strong wind can land many separately thrown coins the same way without the throws having anything to do with each other. Shared handedness is consistent with one first population. It is equally consistent with several, pushed the same direction by the same outside hand. Chirality alone does not settle which.

What can be dated, with real error bars attached, is the population that already had this chemistry locked in. A 2024 study in Nature Ecology & Evolution, led by Moody, Álvarez-Carretero, Mahendrarajah, and Donoghue, puts the last universal common ancestor of all life alive today — LUCA — at approximately 4.2 billion years old: a 95 percent credible interval of 4.09 to 4.33 Gya under the team's main relaxed molecular-clock model, and a full span of 3.94 to 4.52 Gya once every model they compared is counted. The method rests on five pairs of paralogous genes — genes born from a duplication inside one ancient genome, before LUCA existed, then inherited down two separate lines ever after — among them the two subunits of ATP synthase and the two elongation factors, Tu and G, calibrated against thirteen fossil and geochemical tie-points stretching from the Moon-forming impact, 4,510 million years ago, to a 2,954-million-year-old oxygen signature preserved in South Africa's Mozaan Group.

That figure revises an earlier reconstruction: Weiss and colleagues, writing in Nature Microbiology in 2016, described a thermophilic, nitrogen-fixing LUCA built from somewhat different evidence. The newer paper also argues that a ceiling some older studies leaned on — a supposed heavy bombardment of the early Earth by asteroids and comets, around 3.7 to 3.9 Gya — is not a safe upper bound, since the timing, intensity, and even the reality of that bombardment episode is itself disputed in the lunar-impact record. LUCA's date is not an inscription. It is the current best answer, already revised once, open to being revised again.

By the time that population existed, whatever it looked like and wherever it lived, its cells were already reading something: a set of instructions, written in a molecular alphabet, translated then the same way its every descendant translates it now. The choice of hand came first. The choice of letters came next.

[[FIGURE: deep-time-timeline | The road ahead: four billion years on a single page.]]

Overturned if: a lineage of life — ancient, modern, or reconstructed in a laboratory — is shown to run on D-amino acids and L-sugars as its heritable core, rather than as a late, isolated exception, which would mean Earth's uniform handedness reflects one external push acting on separate origins rather than one shared ancestor.

Sources

• 2021 · Vantomme, "Pasteur and chirality: A story of how serendipity favors the prepared minds" — Chirality (Wiley)
• 2016 · "Why are all proteins 'left-handed'? New theory could solve origin of life mystery" — Science/AAAS
• 2011 · "Emerging knowledge of regulatory roles of d-amino acids in bacteria" — Cellular and Molecular Life Sciences (Springer)
• 2006 · Torres et al., "Mammalian L-to-D-amino-acid-residue isomerase from platypus venom" — FEBS Letters
• 2025 · "Life's homochirality: Across a prebiotic network" — PNAS 122
• 2024 · Moody et al., "The nature of the last universal common ancestor and its impact on the early Earth system" — Nature Ecology & Evolution