Grains, not landscape

Earth’s oldest crystals grew before any surviving rock.

Almost no mapped rock survives from Earth’s first 500 million years. The oldest intact rock most workers will defend is about 4.03 billion years old. Older than any outcrop are tiny crystals of zircon, mostly from the Jack Hills of Western Australia, one grain dated 4.404 billion years.

They sit today in conglomerate about 3 billion years old — late Ganymede into Callisto time on the ruler this site already uses. That date is when old grains were eroded and buried. It is not when they grew. The host faces that grew them are gone.

This site reads Io through Mars as consecutive stages of one kind of planet, with living Earth as the present chapter. On that ruler, Io time is 4.54 to 3.97 billion years ago — Earth’s Hadean. The Jack Hills crystals grew then. They grew on Earth. The question is whether a chapter that leaves crystals and almost no landscape is the right kind of chapter to set next to a silicate face that is still being rewritten.

The ordinary view

Zircon is zirconium silicate. It takes uranium, rejects lead when it grows, and is hard enough to outlive granite or volcanic rock. When the host is eroded, melted, or hammered, the crystal can remain. Jack Hills still holds most of the pre-4.0 billion year inventory. Other localities exist. Acasta in Canada is the first defended outcrop, not the first crystal. Nuvvuagittuq ages near 4.16 billion years remain argued.

Four measurements do most of the work. Ages date crystal growth. Oxygen isotopes in many unaltered Hadean grains sit mildly above the mantle range; Valley and colleagues treat that as primary magma chemistry from rock that had already met liquid water — a Cool Early Earth before 4.30 billion years ago. Titanium-in-zircon temperatures cluster near 700 °C, the wet minimum-melt range of granitoids today. Inclusions of quartz and muscovite, if original, point to hydrous melts, not a dry lava lake.

Two clocks in one hill

Equal windows are a modeling choice. They are not a heating law and not a transport law. Growth and burial are different clocks. Mixing them is how a 3.0 billion year bed gets read as a 4.40 billion year landscape.

Table 2. What would move each clock. A new grain inside Io time does not date the conglomerate. A new outcrop of the conglomerate does not date the melt.

A lid that does not last as landscape

Io is the right kind of problem. Its silicate surface is rewritten by lava and sulfur on timescales of years to a million years. Impact craters do not accumulate. Arrive in a billion years and you would not walk an Io outcrop from today. You would need something that survives the reset — a durable mineral later dumped into younger sediment.

That is what Jack Hills is on Earth: first crust as grains, because the host faces were destroyed. The oldest crystal, about 4.40 billion years, sits inside Io time, not at 4.54. About 140 million years of Earth are still missing even as crystals. The Moon-forming impact and the last freeze of a magma ocean sit in that gap. The Io window does not begin at the first zircon.

Tides are the usual source of Io’s heat. They do not rule out heat from the interior as well. Juno argues against a shallow global magma ocean; eruptions can still be fed from a mostly solid mantle into local chambers. Hadean zircons argue a different thermal story on Earth: wet, relatively cool melts after 4.40 billion years, not a surface still floored by a global magma ocean. The analog is reset crust. It is not an identical heat engine, and it does not put Jack Hills granite on Io.

Water arrives late in the first window

Mildly elevated oxygen isotopes before 4.30 billion years ago record water-rock interaction in Earth’s late Hadean — still Io time on the ruler, late in the window. Claims of land and fresh water near 4.0 billion years sit on the Io–Europa edge. That is an Earth chapter. It is not a mapped sea on Io.

Europa’s published state is a global ocean under ice. Earth’s first defended life dates, about 3.7 to 3.5 billion years ago, are already Europa time. Those two sentences stay apart. Acasta at about 4.03 billion years is the first rock pin — the end of Io time. Grains through most of the window. A landscape that lasted at the line.

What this does not detect

It does not put cells on Io or in Europa’s ocean. It does not make Io into the Hadean Earth. Zircon is a survival bias: it records evolved, zirconium-saturated melts and under-samples dry basalt. Io’s visible face is sulfur and basalt. Using grains to describe the whole Hadean already over-weights granite-like melts. The analog inherits that bias. The 4.40 billion year grain is not the birth of continents.

What would weaken this

A defended Hadean landscape still in outcrop would take the grains-not-landscape pairing off the table. A Jack Hills inventory that grew after 3.97 billion years would miss the Io window. A host conglomerate dated inside Io time would collapse growth and burial into one clock. Neither result would prove a handoff.

What this page holds

Earth’s oldest surviving material is mineral, not landscape, in the same interval this site assigns to a silicate face that is still being rewritten. Growth is one clock. Burial is another, a billion years later. This site reads that as archive type on consecutive stages. The record does not prove the handoff.

Sources

Wilde, S. A., Valley, J. W., Peck, W. H., and Graham, C. M. (2001). Jack Hills zircon dated 4.404 ± 0.008 Ga.

Valley, J. W., and colleagues (2002, 2014, 2024). Mildly elevated δ¹⁸O as a primary signature; liquid water before 4.30 Ga.

Watson, E. B., and Harrison, T. M. (2005). Titanium-in-zircon temperatures near 700 °C.

Bowring, S. A., and Williams, I. S. Acasta Gneiss as the oldest widely defended intact rock, about 4.03 Ga.

Park, R. S., and colleagues (2024). Juno gravity: no shallow global magma ocean in Io.