Rockhounding.org
Geode

geode

Mohs 6.5–7

Geode hardness and streak
Geode has a Mohs hardness of 6.5 to 7 and a white streak. It is a gray to white rock. It is mapped at 131 sites in the United States.

rock · common

Field note. Round hollow ball lined with crystals — lighter than a solid rock of its size; rattles sometimes.

Also called geode nodule, hollow nodule, potato rock, keokuk geode.

Interactive lesson

How do geodes form?

A geode starts as a hollow: a gas bubble trapped in lava, or a nodule in limestone. Long after, warm groundwater lines it with agate and then crystals. In the 134-million-year-old basalts of Brazil and Uruguay those crystals are amethyst.

Read this lesson
  1. 1. Lava full of gas bubbles

    Red-hot lava flooded across what is now southern Brazil and Uruguay. Gas fizzed out of it like bubbles in a soda.

    These lava flows are the Paraná basalts. Dating puts them at about 134 million years old, poured out over roughly 1 to 3 million years.

    They cover about 1.2 million km², more than Texas and California together, stacked up to 1,700 m thick.

    Bubbles rose and gathered near the top of each flow. That is why the geodes sit in a band 5 to 9 m thick near the top of a 40 to 60 m flow, and why many stand upright.

    Thiede and Vasconcelos 2010, Geology; Gomes and Vasconcelos 2020; Arduin Rode et al. 2024, Mineralium Deposita.
  2. 2. An empty pocket

    The lava cooled into hard, dark basalt. The biggest bubble was trapped inside as a hollow pocket.

    Most geologists read the pockets as frozen gas bubbles. A smaller group think hot water swelled the rock open later.

    In Uruguay the biggest geodes are up to 5 m long and weigh about 10 tonnes. Upright ones called capillas, chapels, reach 2 to 3 m.

    Proust and Fontaine 2007, Geological Magazine; Gilg et al. 2003, Mineralium Deposita; Duarte et al. 2009; Arduin Rode et al. 2024.
  3. 3. Water seeps in

    Much later, warm groundwater trickled through cracks into the pocket. It carried silica, the stuff quartz is made of. First it left a thin green skin.

    The water was about 15 to 80 °C, closer to a hot bath than to lava. The crystals did not grow in hot lava.

    The silica came from volcanic glass in the basalt, dissolved by water that rose from the aquifer below.

    The green skin is celadonite, a clay mineral about 1 mm thick. We draw it thicker so you can see it. The only dates on it, 90 to 65 million years ago, come from one early study.

    Gilg et al. 2003 and 2014, Mineralium Deposita; Arduin Rode et al. 2024.
  4. 4. Agate lines the walls

    Each trickle of water left a thin layer of silica. Layer after layer, the walls grew bands of agate.

    Agate is chalcedony: quartz in fibres too small to see. It is the most common filling, and many geodes are almost solid agate.

    Some layers lie flat across the bottom of a geode, like water in a glass. They show which way was down when they formed.

    It took a lot of water: at least 163 litres flowed through for every cubic centimetre of quartz left behind.

    Arduin Rode et al. 2024, Mineralium Deposita.
  5. 5. Crystals grow inward

    Then the silica grew as crystals: six-sided quartz points reaching for the middle. Iron in the quartz, zapped by natural radiation in the rock, turned them purple. That is amethyst.

    The crystals grew in water at about 25 to 60 °C in Uruguay and 40 to 80 °C in Brazil.

    Purple comes from iron atoms in the quartz, changed by natural radiation from potassium, uranium and thorium in the rock. Heat amethyst to about 450 °C and it turns yellow: most citrine sold is heated amethyst.

    No one has measured how fast they grew. Our estimate from the water needed is roughly 1 mm every 50,000 to 1,000,000 years.

    Last of all, in many geodes, cooler water below about 45 °C grew white calcite crystals on top of the amethyst.

    Arduin Rode et al. 2024; Gilg et al. 2014; Cheng and Guo 2020, Scientific Reports; GIA.
  6. 6. Out of the rock

    Rain, roots and frost slowly break down the basalt around some geodes. Miners dig the rest straight out of the rock.

    At Ametista do Sul in Brazil the geodes sit in one basalt flow about 40 to 50 m thick, and miners follow that band through the rock.

    A geode is tougher than the rock around it, which is why geodes can weather out whole.

    Gilg et al. 2003, Mineralium Deposita.
  7. 7. Crack it open

    From the outside a geode looks like any lumpy rock. Tap it three times to crack it open.

    Not every round rock is hollow. Many are solid all the way through, and you cannot tell from the outside.

    Bright pink or blue geodes are dyed.

    Geodes in Iowa, Illinois and Missouri formed differently: about 340 million years ago in sea-floor mud, not lava.

    Carnegie Museum of Natural History; Iowa Geological Survey.
  8. 8. You cracked it

    Turn the halves with your finger. Tap any layer to name it.

Check: what turned the crystals purple?

Iron in the quartz, plus natural radiation. Amethyst is quartz with a little iron in it. Natural radiation from the surrounding rock changed that iron, and the changed iron absorbs light in a way that looks purple.

The written version of the interactive lesson “How do geodes form?”.

Interactive lesson

Keokuk geodes: geodes in sedimentary rock

Keokuk geodes in Iowa, Illinois and Missouri formed in Mississippian sea-floor sediments around 340 million years old, where nodules were replaced by quartz and later lined with crystals.

Read this lesson
  1. 1. A warm, shallow sea

    A warm, shallow sea covered what is now Iowa, Illinois and Missouri. Sea lilies and sponges lived on its muddy floor.

    Keokuk geodes come from the lower Warsaw Formation and the top of the Keokuk Formation: shale, shaley dolomite and limestone about 340 million years old, from the Mississippian Period.

    That mud settled on the floor of a shallow inland sea that was slowly draining away. The best collecting is within about 50 miles of Keokuk, Iowa, especially in Lee, Henry and Van Buren counties and across the rivers in Illinois and Missouri.

    We draw the geode much bigger than real compared with the mud around it. Most are fist-sized; the Iowa Geological Survey shows one 8 inches (20 cm) across.

    Iowa Geological Survey; Mahaffey and Finkelman 2022, Minerals 12:914; The Gazette, Cedar Rapids.
  2. 2. A lump grows in the mud

    Just under the sea floor, minerals from the salty water built a hard lump called a nodule. As it grew, it shoved the soft mud layers aside.

    Watch the layers bend around the lump. Real geodes sit in bent and broken layers, which shows the lumps grew while the mud was still soft.

    Geologists disagree about what the lump was made of. Hayes (1964) studied the Warsaw geodes and read them as calcite lumps, the mineral in limestone. Others read them as anhydrite, a mineral like gypsum that grows from very salty water (Chowns and Elkins 1974; Maliva 1987). Bits of anhydrite trapped inside geode shells from the same rock layers in Kentucky and Tennessee support that idea.

    The Warsaw rocks are mostly dolomite, and geodes almost always sit in dolomite. No one has measured how fast the lumps grew.

    Hayes 1964, Journal of Sedimentary Petrology 34:123; Chowns and Elkins 1974, J. Sed. Petrology 44:885; Maliva 1987, J. Sed. Petrology 57:1054; Smith and Siewers 2004, GSA Abstracts 36(3):42; Kentucky Geological Survey.
  3. 3. A shell of silica

    Water creeping through the buried mud carried silica, the stuff quartz is made of. Bit by bit it swapped the outside of the lump for a hard shell.

    The shell is chalcedony: quartz in grains far too small to see. Every Keokuk geode has one.

    No one knows for sure where the silica came from. A strong candidate is sponges, whose skeletons are made of glassy silica needles that dissolve easily. Quartz sand barely dissolves at all.

    The shell grew from the outside in. We draw it thicker than it is so you can see the layers.

    Hayes 1964; Maliva 1987; Smith and Siewers 2004; Kentucky Geological Survey.
  4. 4. The middle dissolves

    Then water dissolved the soft middle of the lump. The tough shell stayed behind with a hollow inside.

    Hayes found every stage, from solid lump to hollow shell, lying side by side in the same rock. That is how the order of the steps was worked out.

    What dissolved the middle is still argued. Hayes thought acid water dissolved the calcite. In the anhydrite idea, the sulfate mineral was dissolved or used up by chemical reactions in the buried water.

    Some lumps never hollowed out. They stay solid all the way through, and from the outside you cannot tell.

    Hayes 1964; Chowns and Elkins 1974; Kentucky Geological Survey.
  5. 5. Crystals grow inward

    Water kept trickling in and left quartz crystals on the inside of the shell, all pointing to the middle. In some geodes calcite and other minerals grew too.

    Most Keokuk geodes are lined with quartz. Many also hold calcite, and some hold dolomite, pyrite, sphalerite, kaolinite, goethite or even a little oil.

    One geode from Hamilton, Illinois, held at least 15 minerals, 5 of them never reported from these geodes before, including tiny grains with gold and silver in the shell.

    The crystals that grew last are the outermost ones you see: here, honey-coloured calcite and brassy pyrite sitting on the quartz.

    Mahaffey and Finkelman 2022, Minerals 12:914.
  6. 6. Out of the rock

    Rain, frost and creeks wore away the soft shale. The tough geodes were left behind in creek beds and at the bottom of slopes.

    Shale crumbles in rain and frost, but a quartz shell hardly wears at all. So geodes collect in streams where the Warsaw Formation is exposed.

    Iowa made the geode its state rock in 1967. Collectors have pulled single geodes over 175 pounds (80 kg) out of the creeks around Keokuk.

    Iowa Geological Survey; The Gazette, Cedar Rapids.
  7. 7. Crack it open

    From the outside a Keokuk geode looks like a bumpy ball of rock. Tap it three times to crack it open.

    A hollow geode feels light for its size, which is how collectors pick the good ones.

    Bright pink or blue geodes in shops are dyed. Keokuk geodes are white, grey, clear and honey brown inside.

    Amethyst geodes from Brazil formed in a different way: in gas bubbles in lava, 134 million years ago.

    Carnegie Museum of Natural History; Iowa Geological Survey.
  8. 8. You cracked it

    Turn the halves with your finger. Tap the shell, the crystals or anything else to name it.

Check: where did Keokuk geodes begin?

As a lump in mud on a shallow sea floor. They began as lumps in sea-floor mud about 340 million years ago. Brazil's amethyst geodes are the ones that began in lava.

The written version of the interactive lesson “Keokuk geodes: geodes in sedimentary rock”.

Interactive lesson

Reading a geode's growth history

The order of layers in a geode wall, from chalcedony to quartz to calcite or other late minerals, records how the fluid's chemistry changed over time.

Read this lesson
  1. 1. Read it from the outside in

    A geode fills from its wall toward the middle, so the layer touching the rock is the oldest. The first one is a thin skin of green clay.

    Each layer sits on the one before it, so geologists put the minerals in order by what grew on what. A team that studied more than 30 mines at Los Catalanes, Uruguay, sorted the filling into three stages: before, during and after the amethyst.

    The first layer is celadonite, a green clay-like mineral about a millimetre thick. We draw every layer much thicker than real.

    Researchers in Brazil found pyrite, fool's gold, in this first stage and read it as water with no free oxygen. They found no trapped drops to measure, so this layer's temperature is unknown.

    Arduin Rode et al. 2024, Mineralium Deposita; Gilg et al. 2014, Mineralium Deposita 49:861.
  2. 2. Agate: cool water, band by band

    Next come bands of agate, quartz in fibres too small to see. The water that left them was only about 15 to 35 °C.

    Oxygen in a crystal comes from the water it grew in, and the mix of heavy and light oxygen depends on how warm that water was. Measured against water still sealed inside the geodes, the agate points to 15 to 35 °C.

    Why fibres and not big crystals? One idea is that water crowded with silica makes it link into chains that crystallise fast and messily. Its author, Peter Heaney, says no lab has shown it yet, so the card marks it with a question mark.

    Arduin Rode et al. 2024, Mineralium Deposita; Penn State Research, "How do agates form?", 2001.
  3. 3. Clear quartz: a little warmer

    Then clear quartz grew, its crystals getting bigger toward the middle. The water probably carried less silica now, and it was a bit warmer.

    Under the microscope, the quartz grains in the Uruguay geodes grow larger toward the hollow middle.

    Trapped drops and oxygen in this quartz give about 30 to 60 °C. The authors say quartz at this stage shows higher temperatures than the agate.

    In Heaney's idea, once the silica chains are used up, single silica molecules build slow, orderly quartz. That part is untested too.

    The same team found that the agate, quartz and amethyst all grew from one changing water, not three different ones.

    Arduin Rode et al. 2024; Penn State Research 2001.
  4. 4. A drop of the old water

    Growing amethyst sealed in tiny drops of the water. Heat one under a microscope, and the moment its bubble vanishes tells you how warm the water was.

    These drops are 1 to 40 micrometres long. Most hold only liquid, no bubble, which already points to cool water near the surface. A drop sealed in hot water splits into liquid and a bubble as it cools.

    To measure one, scientists fire a very short laser pulse to make a bubble, then warm the crystal until the bubble disappears. That temperature is the lowest the water could have been.

    Amethyst in Uruguay gives 15 to 55 °C. We show one drop clearing at 35 °C as an example. Drops that already had bubbles gave 144 to 385 °C, but they had leaked, so the authors set them aside.

    Arduin Rode et al. 2024; Université de Lille fluid inclusion course; McBride 2019, Gems & Gemology 55(1).
  5. 5. Calcite comes last

    Last, white calcite crystals grew on the amethyst tips. Calcite is made of calcium and carbonate, not silica, so the water had changed again.

    The Uruguay team links the silica layers to fast pulses of water, and calcite to slower water that soaked longer in the basalt. Late calcite there grew at about 17 to 55 °C.

    In Brazil, calcite and the gypsum after it grew below 45 °C, cooler than the amethyst at 40 to 80 °C.

    The Brazilian drops also get fresher from early to late: about 5.3, then 3.4, then 0.3 % salt. Seawater is about 3.5 %, so the water went from half again as salty as the sea to about a twelfth as salty.

    Arduin Rode et al. 2024; Gilg et al. 2014; USGS, Why is the ocean salty?
  6. 6. Two places, two stories

    Geologists read the same kind of geode in Brazil and in Uruguay and got different stories. In Brazil the water looks warmer and saltier; in Uruguay it looks like ordinary groundwater.

    Ametista do Sul, Brazil: quartz and amethyst at 40 to 80 °C, from pulses of water up to 80 to 90 °C entering rock below 45 °C. The authors think it was salty water from deep in the Guaraní aquifer.

    Los Catalanes, Uruguay: 15 to 60 °C overall, never more than about 3 % salt, and no steady trend. The authors read it as rain-fed groundwater, and say Brazil's high salt and steady trend do not show up there.

    A 2007 study read 152 to 238 °C from drops in Brazilian amethyst and argued for heat from the volcano. The Uruguay team says drops like those had leaked. Both recent studies agree the water was far below boiling.

    No one knows how long the layers took. The only dates, 90 to 65 million years ago, are on the green clay from Brazil and come from one early study.

    Gilg et al. 2014, Mineralium Deposita; Arduin Rode et al. 2024, Mineralium Deposita; Proust and Fontaine 2007, Geological Magazine.
  7. 7. Put it in order

    Now you read it. Tap the five layers from oldest to newest.

    Look for what sits on what: a layer that grew over another is younger than it.

    Real geodes are not always this tidy. In some, agate, quartz and amethyst repeat in turn, a record of water arriving in pulses.

    Arduin Rode et al. 2024, Mineralium Deposita.
  8. 8. You read the wall

    Clay, agate, quartz, amethyst, calcite: oldest to newest. Tap any layer to look at it through the lens.

Check: how can scientists tell how warm water was millions of years ago?

From tiny drops of it sealed inside the crystals. Crystals sealed in tiny drops of the water as they grew. Heat a drop until its bubble vanishes and you have a temperature. The oxygen in the crystals gives a second check.

The written version of the interactive lesson “Reading a geode's growth history”.

Home tests

How to identify geode

The tell: Round + lighter than it looks: cut or crack it.

StreakUnglazed tile

Drag it across the back of a ceramic tile. Geode leaves a white or no streak.

HardnessCopper coin

A copper coin (3.5) does not mark it. A steel blade (5.5) slides off, and it scratches glass.

VinegarHousehold vinegar

Vinegar does nothing. Geode is not a carbonate.

UltravioletUV flashlight

Geode does not fluoresce under UV.

Drawinggeode
drag to turn · move the light

A drawing made from its color and luster.

Send photos so we can scan one →

How hard is geode?

Geode is 6.5 to 7 on the Mohs scale. A copper coin (3.5) does not mark it. A steel blade (5.5) slides off, and it scratches glass.

1
2
3
4
5
6
7
8
9
10
Geode 6.5–7

Mohs scale from 1, talc, to 10, diamond. A fingernail is 2.5, a copper coin 3.5, a steel knife 5.5 and a steel file 6.5.

Geode properties
Mohs hardness
6.5–7A copper coin does not mark it; a steel blade slides off.
Streak
white
Color
graywhite
Luster
glassy
Light
opaque
Habit
botryoidal
Fracture
conchoidal, uneven
In vinegar
No reaction
Magnetic
No
Data: Rock ID catalogue and IMA data
Where geode is found
131 sites in the United States

From mine records and member reports on the rockhounding.org map.

Open the geode map
How hard is geode?

Geode has a Mohs hardness of 6.5–7. A copper coin (3.5) does not mark it. A steel blade (5.5) slides off, and it scratches glass.

What color is the streak of geode?

Geode leaves a white streak on an unglazed ceramic tile. The streak is the color of the powder and is more reliable than the color of the surface.

Does geode react with vinegar?

Vinegar does nothing. Geode is not a carbonate.

Where is geode found?

Geode is recorded at 131 sites on the rockhounding.org map, 131 of them in the United States. The most are in Iowa (16), Kentucky (11) and Illinois (11).

Where can I sell geode?

Unopened geodes sell as a gamble, and opened ones sell on how the inside looks: halves lined with even, undamaged crystals bring the most. A matched pair of halves sells for more than a single half. You can sell it at a rock shop, a gem and mineral show or a local club, or online on eBay, Etsy or the rockhounding.org market.

What is geode confused with?

Chert, granite, diorite and andesite. Geode is glassy; chert is waxy.

Photographs of geode

Geode
Photo: CEphoto, Uwe Aranas · CC BY-SA 4.0, Wikimedia Commons
No real scan yet

We have no 3D scan of geode yet. If you own a good piece, send phone photos and we will build one, credited to you.

Photo credits, Wikimedia Commons: Chert by James St. John, CC BY 2.0; Granite by James St. John, CC BY 2.0; Diorite by USGS, Public domain; Andesite by Smithsonian Institution, CC0.

Rockhounding.org editorialProperties from the Rock ID catalogue and IMA data