Keokuk Geodes

Keokuk, Iowa is situated at the confluence of the Mississippi and Des Moines Rivers in the extreme southeastern portion of the state of Iowa. Illinois borders Iowa on the east and Missouri is to the south (area known locally as the "Tri-State area").  See map below for a visual view:

Map Credit/Reference:  City Data (

This "information only" page (none of the specimens pictured is for sale) provides a non-technical approach to Keokuk Geodes.  Below is a general description of Keokuk geodes, the minerals found in the geodes (many with pictures), and a section on credits, general facts, and personal observations/opinions, as well as important citations (by no means a comprehensive list - just what was used here specifically).  Please continue to come back for additional descriptions, updates, and pictures.  Current work is focused on geode mineral photos, geodes "in situ" (in their host rock on site before removal), mineral deposition/growth and related paragenesis in the cavity with respect to gravity, and geode formation.  August 2011 update:  Please note that a major overhaul to this page will be taking place by the end of the year and there may be some down time and broken links as we make the transition to a much more modern and organized look to this web page.  Some new, larger, clearer pictures to assist in identification have been added below.

Keokuk, Iowa is known as "the geode capital of the world" for very good reason. A virtually countless number of geodes has been collected from the Keokuk area over especially the last 150 years. There are no less than 300 outcrops of the Lower Warsaw Formation that the geodes are found in within a 100 mile radius of the city. The main exposures are found in streambeds that are tributaries of the Mississippi River. (Sinotte, 1969; Hess et al. 1998). The two predominant types of rock in the Lower Warsaw Formation are argillaceous dolomite and dolomitic mudstone.  The geodes found loose in creek beds originated in the Lower Warsaw Formation, but over time weathered out of the rock and fell into the creek, with many being transported downstream away from their source.  In some locations where the Lower Warsaw Formation is right along the creek bank or even in the creek itself, it is advantageous to collect geodes right after a heavy rain that may cause a rise in both the creek level and the water current as fresh geodes may be loosened or dislodged from their host rock due to the faster-moving water (for your safety, wait until the water goes back down, however!).  Pictured below are two opened geodes from the Keokuk area in their host rock, with close-ups of each piece.  Below that is a picture of an un-opened geode in its host rock.

Click on Images For Larger View

Figures 1a-c:  Opened Keokuk geodes in their host matrix.  The two matrix pieces (including the geodes) fit back together so the geodes are complete.  Each matrix specimen measures approximately 6-8 inches across.  The geodes are lined with quartz and calcite.

Figure 2:  Un-opened Keokuk geode in its host matrix.   Even though both the geode and the matrix are the same color, the matrix is usually flat or in layers while the geode is round, so geodes appear as "bubbles" in the matrix and look out of place.  The geode shown here is approximately 2.5 inches in diameter.

  Minerals in Keokuk Geodes (Primarily from the Warsaw Formation)

The following minerals have been positively identified as being in Keokuk geodes, as well as those previously published to exist or are currently under study.  The photos for each mineral are currently under construction, please check back from time-to-time to see new additions.

Quartz- Lines almost all geodes. Clear, smoky, sea-green, iron-stained (orange), and pink or cherry tinted geodes are found (see photos below).  Clear and iron-stained geodes are the most commonly seen type of quartz geode.  To date, research findings have not confirmed the presence of "citrine" geodes in the Keokuk, Iowa area.  All such samples have been identified as iron-stained geodes.  Citrine is considered by many geologists and gemologists to be rather uncommon or infrequent anywhere on Earth, and is considered by many in the gemological sector to be more uncommon than amethyst (Holden, 1999).  Many iron-stained Keokuk geodes can be cleaned with any iron-removing product or various acids.  In the samples where the orange/yellow coloring does not clean out, it was found that the samples had successive layers of quartz on top of existing iron-stained quartz, but due to the transparency of the quartz crystals, the orange color still showed through.  Many geode samples have also been identified with "rings" of pyrite in the geode rind, indicating successive growth stages of quartz, then pyrite, then quartz again, etc.  Research by Sinotte (1969) also labels all orange quartz geodes as iron-stained.  The iron-staining is caused by the weathering of various sulfide minerals, most notably pyrite but also marcasite, chalcopyrite, etc.   The term "citrine colored" quartz is appropriate when describing Keokuk geodes, but true "citrine geodes" do not occur in the Keokuk area.  The pink or cherry tinted geodes, if naturally occurring (recently, several so called "red quartz geodes" were made available but they were simply orange iron-stained geodes that had been baked in an oven to a certain temperature to turn the orange to a red color) are occasionally caused by a close association with the surrounding host rock or where the chalcedony lining contained excessive iron amounts and resulted in a more of a reddish tint in the geodes. These will also clean up unless having been under the process of weathering for an extended period of time or subsequent growths of quartz have overlaid the layers of red iron deposits. In other geodes, the faces of several quartz crystals are coated with a very weak pink to vivid red, and can be washed out easily, these are most likely associated with hematite (see hematite section below). At most locations where hematite is the reason for the reddish coloring, other geodes in the same exposure will have actual hematite crystals and geodes containing pink to red water have been collected at a few locations.


Figure 3a, b (two left photos):  Typical Quartz-Lined Keokuk Geode.  Each half measures approximately 4 inches in diameter.

Figure 3 c, d (two right photos):  Smoky Quartz-Lined Keokuk Geode.  Each half measures approximately 10 inches in diameter.  Note the considerably darker crystals in the right two photos versus the left two photos.


Figure 3 e, f (two left photos):  Light sea-green Quartz-Lined Keokuk Geode.  Each half measures approximately 6 inches in diameter.  This particular type of geode was quite abundant at the original St. Francisville Mine when it was open and accessible, and many such geodes from this location often contained large clusters of pink calcite crystals making for gorgeous specimens.

Figure 3 g, h (two right photos):  Iron-stained Quartz-Lined Keokuk Geode.  Each half measures approximately 3.25 inches in diameter.  This particular geode also contains very dull and incomplete pyrite crystals, further indicating the presence of weathering.

"Snowball" Geodes:  An interesting variety of mainly quartz geode that is well-known among Keokuk geode collectors is the "snowball" geode.  Sinotte (1969) apparently first coined the term, and while definitions vary as to the true definition of a "snowball" geode from collector to collector, the main requirement for a Keokuk geode to be labeled as a "snowball" geode is the presence (or at least former presence) of a silica concretion in the interior of the geode (Sinotte, 1969 - refer to pages 74-77) that was later coated over by quartz and sometimes additional minerals, forming an attached round ball of quartz to the interior of the geode cavity.  Objectively defined, if there is proof of a separate entity in the form of the silica concretion in the geode cavity, regardless of shape, separate from the lining of the geode to at least some degree, it is, by definition, a "snowball" geode.  Some samples are formed better than others and contain less connections to the quartz wall and a more pronounced "ball", but the basic definition is still present in the more poorly or better-developed specimens.  Several such samples have been collected that noticeably contain such "snowballs", but the interior silica/chalcedony (usually a white solid material that is commonly mislabeled as kaolinite due to its appearance) has since disappeared, leaving an open void except for the layer of quartz on the outside.   The author of this page has collected several samples and also studied several other samples where this has been observed.  Such a phenomena may indicate an instability in the former interior silica ball in relation to the surrounding quartz, and a dissolving of the silica may have occurred after the silica ball was coated with quartz.  Research is ongoing - if anyone else has samples that fit this description or a sound theory based on research studies related to this topic, please contact me with your observations/opinions or references to the process that may have occurred and if possible, pictures of the samples under study or in question.  Below are two geodes containing "snowballs" - the left geode shows a hollow snowball and the right is a regular, typical snowball. 


Figures 3 i, j (two left photos):  Quartz (hollow "snowball") Keokuk geode - Note the left half and closeup of the left half - the geode has the standard outside lining but also appears to have a geode within a geode.  When opened, the snowball on the inside also opened, exposing the hollow space in the middle.  An explanation for this is not currently available in the literature, but may be related to the dissolving of the silica solution on the inside of the snowball, leaving only the crystallized quartz on the outside of the former snowball interior because the silica was more apt to dissolving but the crystallized quartz was not.  The larger half is approximately 3 inches across.

Figures 3 k, l (two right photos):  Quartz "snowball" Keokuk geode with secondaries of Calcite, Pyrite.  The "snowball" is the spherical cluster of quartz seen in the just left of center of the left photo, with a closeup in the right photo.  The geode section is also graced with the presence of calcites and pyrites below the snowball.  The geode itself is approximately 8 inches up and down, with the snowball being approximately 3 inches in diameter.


Chalcedony (Quartz)- This non or micro-crystalline variety of quartz is found mostly as being blue or grey, but also orange, black, red, brown, and green chalcedony is found. It appears as mammillary (bubbles, or "brains"), the base lining after quartz below the quartzoid crystals in dewdrop diamond geodes (photo below), and thin coatings over quartz that gives square-like crystal habits in many geodes.


Figures 4a, b (two left photos): "Dewdrop Diamond" Quartz Keokuk geode with secondary calcite in the right half.  The white band around the edge of the geode is the white chalcedony, followed by clear quartzoid crystals (see closeup for a better view).  The crystal mass in the bottom of the right half is calcite.  Sometimes just a few quartzoid crystals are found on the white chalcedony, but usually the majority of the white chalcedony is coated with quartzoid crystals that sparkle like dew drops (hence the nickname).  The quartzoid crystals can also be smoky in color (see photo below).  Each half of the geode here is approximately 3.5 inches in diameter.  Sinotte (1969)  lists dewdrop geodes occurring at a frequency of approximately 1:10,000 Keokuk geodes, but at a few select sites dewdrop diamond geodes are quite plentiful in certain horizons of the formation (locally abundant).

Figures 4c, d (two right photos):  Quartz (Chalcedony) Keokuk geode.  The orange color in the bottom of the right half is iron-staining caused by the weathering of most likely marcasite in this particular sample.  The chalcedony in this particular geode is a blue/grey, but it can also be many other colors.  Each half of this pair is approximately 2 inches in diameter.

Figures 4 e, f:  Smoky "Dewdrop Diamond" Quartz Keokuk geode.  Each half is approximately 2.25 inches in diameter.  Note the presence of the white chalcedony band in between the outlying quartz rind and the smoky quartz crystals in the middle.  Sinotte (1969) observed that the thicker this white chalcedony was, the larger the quartzoid crystals on the inside. 


Calcite- Probably one of the prettiest minerals and comes in the most varied crystal structures found in Keokuk geodes. It comes in white, colorless, pink, black, brown, and yellow colors. Many forms of calcite exist, such as the basic rhombohedron, scalenohedron, barrel (prismatic), tabular, and many other more scientifically oriented terms with modifications to the primary crystal type and also varying forms of twinning at some sites. Several different stages of crystallization may occur in the same geode making for beautiful specimens.

Photographs of Calcite in Keokuk Geodes:


Figures 5a, b (two left photos):  Calcite lined Keokuk geode with a secondary "nailhead" or "pagoda" calcite crystal.  The nailhead crystal is actually a scalenohedral (dogtooth) calcite crystal that has been overgrown with rhombohedral calcite crystals.  The nailhead calcite crystal shown here is approximately 2 inches tall, with the geode having a diameter up and down of approximately 3 inches.

Figures 5c, d (two right photos):  Chalcedony (Quartz) lined Keokuk geode with a secondary rhombohedral calcite crystal in the right half.  The calcite crystal shown here in the right half is approximately 3 inches across and shows striations on the crystal face.  The geode itself is approximately 5 inches in diameter.


Figures 5e, f (two left photos):  Quartz lined Keokuk geode with a secondary calcite in the right half.  The geode itself is approximately 7 inches in diameter, with the calcite 1.75-2 inches across. 

Figures 5g, h (two right photos):  Calcite lined Keokuk geode - this geode contains a lining of brown iridescent calcite with secondary clear/white calcite crystals.  Secondary crystals of clear/white calcite are commonly found on the initial brown calcite linings in Keokuk geodes.  Each half of this geode is approximately 3.5 inches in diameter.


Figures 5i, j (two left photos):  Pink Calcite in a Chalcedony lined Keokuk geode.  Calcite on Chalcedony in geodes is frequently seen at many locations in the Keokuk area.  The larger (right half) of this pair is 5 inches in diameter.

Figures 5k, l (two right photos):  Pink Calcite "nailhead" or "pagoda" with White/Clear Calcite, on a Brown Calcite lined Keokuk geode.  In addition to secondaries of white calcite on brown calcite,  pink calcite on brown calcite is also frequently observed.  The pink calcite pagoda here is approximately 2 1/4 inches long, with each half of the geode being approximately 3.25 inches in diameter.

Figure 5m:  Scalenohedral or "dogtooth" Calcite in a Keokuk geode.  Dogtooth calcites can also be white, clear, brown, or even black.  It is not uncommon to find calcite habit and color change in Keokuk geodes so there may be more than one type of calcite in the same geode.


Pyrite- Pyrite is found as black, golden, or reddish crystals in geodes. They are found as being primarily cubic in nature and little more than specks in appearance, but also capillary crystals to 3" occur at some sites, and octahedral crystals occur in addition to modifications of these basic crystal types.  Some geodes contain so many pyrite crystals it appears the geode is lined with pyrite versus quartz or calcite.  Pyrite is also frequently found as clusters of crystals scattered about in the lower Warsaw Formation and as a coating or crystal cluster on top of geodes.  Excellent clusters and lines of pyrite several inches long were frequently found in the "old mine" at Sheffler's and also at the Canton, Missouri roadcut, but they also exist at many other locations as well.

Figures 6a, b, c:  Pyrite crystals on quartz in a Keokuk geode.  The right half of this geode pair contains a heavy concentration of cubic golden pyrite crystals.  Each half of the geode is approximately 3 1/4 inches in diameter.

Figures 6 d, e:  Pyrite crystals and cluster on a chalcedony-lined Keokuk geode.  The geode itself is 4 inches in diameter, with several scattered pyrite crystals throughout the entire geode.  The left half has a 5/8 inch across cluster of above-average sized inter-grown pyrite (click on photo for closeup).

Figure 6f:  Filiform pyrite crystals on quartz in a Keokuk geode.  Click on image for a larger view.  Note the sharp right angle growths in the crystals.  No other mineral in Keokuk geodes exhibits such a growth habit so this helps separate filiform pyrite from marcasite.


Marcasite- This is an interesting mineral commonly mistaken to be pyrite, but there is a difference collectors! Marcasite is one of the most commonly mislabeled minerals in many exhibits of geodes.  Marcasite comes in one form, the capillary crystal. The difference between the pyrite and marcasite is marcasite has protrusions from the main crystal, referred to as resembling a TV antennae (Hess et al. 1998) and also appear as sword-shaped crystals but with very small or no protrusions (different crystal structure).  Magnification is not always necessary but recommended to discern between marcasite and pyrite.  So far, millerite has not been confirmed in Warsaw Formation geodes from the Keokuk area - these geodes are usually pyrite or marcasite-bearing (Sinotte 1969, Hess et al., 1998). Marcasite is quite common at a number of locations but is frequently mislabled as pyrite or vice-versa.


Figures 7a, b:  Marcasite crystals in a chalcedony (quartz) lined Keokuk geode.  The TV antennae crystal shape is especially evident in the photograph at right  - all of the marcasite crystals in the geode at left look just like the crystals in the right image (zoomed in).


Gypsum ("Selenite")- Gypsum is a lesser found mineral than the ones listed above, but the "old mine" at Sheffler's was probably one of the best sites for finding this mineral especially in the very top layer of geodes in the mine (brown skins) and also in the far west end of the pit but it is also found at scattered exposures along or near the Fox River in Clark County, Missouri and was occasionally found in the US 61 Canton, MO roadcut in quartz and brown calcite geodes in addition to widely scattered locations throughout the Keokuk area. These crystals are clear, rod-like crystals up to 2" in length. The difference between gypsum and aragonite (discussed next) is gypsum has a monoclinic crystal structure and is beveled or almost frayed edge at the termination, versus the aragonite which ends at a point.  Barite crystals may also resemble gypsum, but have a different termination than gypsum (orthorhombic).  See Sinotte (1969) for good pictures of the difference in terminations among gypsum, aragonite, and barite.

Figure 8:  Gypsum ("selenite") crystals on quartz in a Keokuk geode.  The crystals are beveled at their terminations and measure up to 2cm (0.8 inches) in length.  Each half of the geode itself is approximately 5 inches in diameter.


Aragonite- Aragonite is not found on a widespread basis in Keokuk geodes, but presently is locally abundant at specific sites. It is a clear, rod-like crystal that resembles gypsum ("selenite"), but ends at a point at the top termination.  Usually associated with ferroan dolomite but not always, and in our experience, makes up a sizable percentage of the interior mineralization in terms of coverage vs. just a single cluster or few single crystals as in gypsum.  Aragonite, in our experience, forms its best and heaviest clusters of aragonite in more poorly formed geodes or geodes with a weaker, thinner skin.

Figure 9 a-c:  Aragonite crystals along with ferroan dolomite and kaolinite on a reddish quartz lined geode.  Due to the small crystal size the photographs are not the best at depicting the aragonite really closely, but do note the difference in crystal structure between gypsum (above) and the aragonite in this geode.  The entire geode measures a little over 3 inches in diameter.  Note the relatively high coverage of aragonite in the geode and the association with the brown ferroan dolomite.  We have observed a close relationship (but not always) between aragonite and ferroan dolomite in nearly all of the geodes with a higher concentration of aragonite. 


Barite- Barite comes in blue and yellow crystal colors, and are mostly tabular in nature, with some groupings of tabular crystals forming rosettes of barite are also common.  Most barite crystals are an inch or less across an edge, but some exceptionally large and well-formed crystals (for Keokuk geodes) up to and over 2 inches have been found recently.  Usually, but not always, the larger crystals are yellowish in color, with the smaller crystals being a white or blue (caused presumably by kaolinite that stunted the growth of the crystal - Sinotte (1969)).  Some crystals are nearly white due to extreme ingestion of kaolinite. Barite is very common at a number of locations especially in Hancock County, Illinois, and was found in the western sections of the "old mine" on Sheffler's property, primarily as tan crystals, some with a degree of kaolinite ingested giving a whitish hue to some or all of the crystal.  The original St. Francisville Mine was also noted for yellow barite crystals.  Barite crystals can also be found as similarly-appearing gypsum ("selenite") crystals, but the termination is different.


Figure 10a (left photo):  Barite and calcite on quartz in a Keokuk geode.  The barite is the yellow tabular crystal at center.  Note that this particular barite has edges that are slightly white - this is caused by kaolinite.  The dark spots on the barite are pyrite crystals.  The barite shown here is approximately 2cm (0.8 inches) across.

Figure 10b, c (two right photos):  Barite, Kaolinite, Calcite in a Chalcedony lined Keokuk geode.  The barites are the two yellow crystals in the left half (and in the right closeup photograph), the kaolinite is the white substance in the right half, and the calcites are the dark brown crystals at the left rear of the right half).   The front barite crystal measures just over 1 inch across, while the back crystal measures 1.6 inches across (from left to right).  The diameter of the right half of the geode pair is approximately 3.25 inches.

Figures 10d:  (left) Barite and Kaolinite on Quartz in a Keokuk geode.  The barite is the cluster of blue, tabular crystals at the center of the geode.  The kaolinite is the white substance especially concentrated to the right of the barite.  The barites in this particular geode are about 1 cm (0.4 inches) tall.

Figure 10e:  (center) Barite with a sharp white Kaolinite edge associated with Pyrite and Quartz in a Keokuk geode.  In locations where yellow barite is the norm vs. blue barite, the crystals frequently have a varying degree of kaolinite on their upper terminations.  Tabular barite crystals are uniquely shaped compared to other Keokuk geode minerals so this helps with identification.

Figure 10f:  (right) Barite with lighter white Kaolinite edge in a quartz lined Keokuk geode.  The yellow barite measures about 1.60 inches across and is found in an 8 inch diameter quartz lined Keokuk geode.  Note the tabular shape to the crystal as well as the color - the yellow color and tabular crystal shape are both dead giveaways that this is a barite crystal and not calcite which may also occasionally be platy or thin but usually with a rhombohedral structure.


Sphalerite- Found in geodes as black crystals up to 4 inches across in some geodes, but are typically found as massive crystals, especially in mudders (concretions) or flattened geodes. Some crystals have smithsonite either decomposed or in small crystalline form (see "Smithsonite" section below), or pyrite can be found on the faces of crystals.  Occasionally, yellow barite was found in close association with sphalerite in the western sections of the "old mine" at Sheffler's.  Sphalerite is quite common in geodes from specific sites in the Keokuk area, especially those locations where the upper sub-unit of the lower section of the Warsaw Formation is observed and especially certain locations where higher zinc contents are present in the overlying bedrock (Sinotte, 1969).  Exceptional sphalerites up to 2 inches have been found recently at a location in Hancock County, Illinois where the cavity was of sufficient size that the sphalerite did not overwhelm the cavity.  Additionally, where the lower Warsaw Formation comes in close association with or is intertwined with the Keokuk Formation, some large sphalerites (up to 4 inches across) are found in poorly developed geodes and/or concretions.  In our experience, the best sphalerite specimens are found in poorly-formed or partially crushed geodes, or concretions with cavities but some well-formed geodes with sphalerites are also present at some locations but the crystal size is, on average, smaller in the well-formed geodes.  Pyrite and/or chalcopyrite crystals are frequently observed on the faces of sphalerites.  If you are unsure if the overgrowth on sphalerite is pyrite or chalcopyrite, look for evidence of greenish colored malachite stains in association with the crystals.  If malachite is present, you have chalcopyrite.  Otherwise you need magnification to confirm pyrite or chalcopyrite.

Figures 11a, b:  Sphalerite on quartz in a Keokuk geode.  Almost without exception kaolinite is associated with sphalerite in Keokuk geodes, and kaolinite (white mineral) is present immediately to the left of the sphalerite and can be seen in the right close-up photo.  The geode has a diameter of approximately 3 inches with the sphalerite being around 1 inch across.

Figure 11c:  Sphalerite associated with Calcite in a Keokuk geode.  The sphalerite crystal here is about 1.25 inches across.

Also see the "Smithsonite" section below for an additional specimen of sphalerite with smithsonite crystals.


Kaolinite- Kaolinite is always found with sphalerite to some degree, but does occur by itself in many geodes. It is a white, powdery mineral, also associated with calcite crystals. Some geodes contain calcites that have ingested kaolinite or are pseudomorphs of kaolinite after calcite. These were quite common at the "old mine" on Sheffler's property in the northern section of the mine, and are found at scattered locations in the Keokuk area, especially in Clark County, Missouri and in the Nauvoo and Hamilton, Illinois area.  Kaolinite has been observed to be a "key player" in the formation/alteration of several other minerals in Keokuk geodes, notably changing crystal habits/colors in calcite, altering the color of sphalerite and ferroan dolomite, and minimizing growth and changing the color of barite. 

Figure 12:  Kaolinite in a Calcite-Lined Keokuk geode.  Field of view is about 3 inches across.  Kaolinite is frequently observed in geodes in the Upper Sub-Unit of the Lower Warsaw Formation, and is almost always, without exception, a companion in a sphalerite-bearing geode (Sinotte, 1969).


Pyrolusite- This is found under the microscope in crystalline form, but also as very iridescent rods at Sheffler's Mine (unconfirmed and questionable). It also makes some chalcedony geodes jet black in color (decomposition - Sinotte 1969).  Research of pyrolusite in Keokuk geodes is ongoing due to the lack of quality and also many questionable samples containing this mineral.  The freqnency of pyrolusite in Keokuk geodes may be more regular than the mineral is given credit for given the lack of good crystallization and undesirable coloring it can impart on geode linings.

Figure 35 a, b:  Black dendritic pyrolusite on quartz in a Keokuk geode.  A large closeup of the right image show the black dendritic growths of pyrolusite.  This is probably a mineral that is more common than is led to believe just because of the lack of really good samples and poor crystal definition to the naked eye.


Goethite- This is a very unique mineral forming 1/2" maximum rods in quartz lined geodes, jet black in nature. Also found as orange stains in geodes, especially at locations where goethite occurs (decomposition - Sinotte 1969).  Goethite is not widespread in the Keokuk area, and is known at only a handful of sites, but is quite common at the sites that it does occur.  At sites where goethite is absent in geodes, it is very likely that the orange coloring in the geodes is caused by weathered pyrite, marcasite, and/or chalcopyrite, not goethite.

Figures 13a, b:  Goethite crystals on quartz in a Keokuk geode.  From a distance (like shown here before expanding picture size), the goethite crystals look like spots of pyrite, but upon closer inspection (click on middle and right photos) the crystals are black rod-like crystals.  While similar in appearance to pyrite or marcasite rods from a distance, upon closer inspection there is an obvious difference in crystal structure.


Hematite- Hematite is the product that causes pink to cherry tinted quartz geodes (see above in quartz section), but also found as red or brownish granular substances as well as small, blood colored crystals in geodes.  A very common mineral presently at select locations in the Keokuk area.

Figures 14a, b:  Red tinting caused by hematite on quartz in a Keokuk geode.  Note the incomplete coverage of the red coating - there are some areas that have clear quartz and no tinting (the red occurs where hematite was present in crystalline form previously).  If water was applied to the interior of this geode, the red coating would easily wash out, indicating a rather recent weathering of the hematite on the faces of the quartz crystals.  Some geodes have been opened at locations where these geodes have been found that have released RED water, not clear water.  This water is currently under analysis to further confirm the presence of hematite.


Smithsonite- Smithsonite is found as crystals on sphalerite or an odd color on sphalerite faces due to weathering.  The crystals or weathering product are usually a brown, tan, or peach color.  Well-developed smithsonite crystals were found at the Highway 61 roadcut near Canton, Missouri on sphalerite crystals but it is also found at nearly all other locations where sphalerites are encountered.  Some exceptional smithsonite crystals are found at a couple locations in Hancock County, Illinois where sphalerites are commonly encountered - one location is Jacob's Mine.  The sphalerites/smithsonites from this location are not always the most aesthetic but do show excellent mineral paragensis.

Figures 15 a-c:  Smithsonite ON sphalerite in association with ferroan dolomite on a Keokuk quartz geode section.  The smithsonite are the two peach/tan colored crystals toward the top right portion of the sphalerite crystal in the MIDDLE photograph on the portion of the sphalerite crystal that has turned a brownish/olive color due to inclusion of kaolinite.  They are small crystals but are visible if you look really close.  The sphalerite crystal is approximately 1.5 inches across.  Note the close association between kaolinite (white powerdy mineral shown in the left photographs) and sphalerite in this geode as well.


Dolomite/Ferroan Dolomite- Very interesting saddle shaped crystals in white, brown, orange, pink, yellow, and black crystals occur. Ankerite has not been positively identified in Keokuk geodes (Sinotte 1969) despite previous/recent published works due to the lack of enough iron in the dolomite (even in very dark dolomite crystals).  Research continues on the possibility of ankerite in some Keokuk geode samples.

Figures 16 a, b, c:  Ferroan Dolomite on Quartz in a Keokuk geode.  Each half is approximately 3 inches in diameter.  Note the "saddle-shaped" crystals at far right.  Ferroan dolomite geodes are currently especially abundant at many sites in especially Hancock County, Illinois but is noticeably less-frequent (except at specific sites) in Clark County, Missouri, possibly owing to a decrease in elevation of 50-75 feet in Missouri and a lack of key constituents in the bedrock.  The geode above with dolomite is just  one color observed - ferroan dolomite can also be pink, brown, very dark brown, reddish, and even close to a black color in certain specimens.


Chalcopyrite- This mineral is found as very small or microscopic crystals in golden form. It has been found at many locations in Hancock County, Illinois, especially in the Warsaw and Nauvoo area, and at a few locations in Clark County, Missouri.  Frequently malachite is found in association with the chalcopyrite as a weathering product and the presence of malachite (greenish crystals or coatings) is a dead giveaway that the iron mineral in the geode is chalcopyrite.  Weathered pyrite and marcasite do not produce malachite as a weathering product.  Crystals are rather small, so identification is best seen under at least 10x magnification.

Figure 17a, b - Crystals of chalcopyrite altering to malachite on chalcedony in a Keokuk geode.  Note the triangular nature and greenish color to the crystals (especially the crystal at far right).  These crystals are only a few millimeters across.  The right photo exhibits malachite on chalcopyrite in a zoomed closeup image of another geode.  Note the close association between the malachite and chalcopyrite indicating a weathering product.


Jarosite- Actually reported by Sinotte (1969) to be a very common mineral in geodes found as microscopic crystals, but mostly found as a granular powder or stains appearing yellow in especially chalcedony geodes.  Due to the lack of larger crystals visible to the naked eye, jarosite is not well-known but again is likely much more frequent than currently given credit for.  If someone has a confirmed specimen of jarosite in a Keokuk geode (submitted for analysis) and you have a picture of it, please contact me for inclusion in this section.


Malachite- Malachite is found as green crystals in geodes that contain quartz and/or brown calcite linings (weathering product), and as crystals/coatings on chalcopyrite crystals. The author of this page has found malachite-bearing geodes in association with confirmed millerite and polydymite in chalcedony lined geodes from the Upper Keokuk Formation at an exposure in the vicinity of Hamilton, Illinois.  Other malachite-bearing geodes have been found at scattered locations in the Dallas City, Niota, Nauvoo, and the Hamilton/Warsaw area of Hancock County, Illinois, and in Clark County, Missouri in association with chalcopyrite crystals.

Figure 18:  Malachite heavily encrusting weathering chalcopyrite in a calcite lined Keokuk geode pair.  Malachite is not difficult to identify due to its green color and association with chalcopyrite crystals.  The malachite on chalcopyrite in this picture is approximately 1 cm across (0.4 inches).


Stilpnosiderite- This "mineral" has been reported to be responsible for producing the peacock iridescent hues in brown calcite lined geodes. Some question has been raised recently, however, if stilpnosiderite is actually possible or responsible for the iridescent hues in Keokuk geodes or if just plain iron in/on the calcite is responsible. Further research is required and ongoing.  The mineral name is also questionable for acceptance as its own species (presentation given at the October 2003 geode symposium in Keokuk, Iowa).  Some locations in the Keokuk area produce very intense iridescent coatings on the faces of non-brown calcite crystals in quartz and calcite lined geodes, and these are, for sure, just an iron coating or a mix of small, micro pyrite crystals and weathered pyrite.  This supposed "mineral" needs more research and it is recommended that any labels make note of the questionable status of this mineral, if included on the label.


Other Substances Encountered in Keokuk Geodes

Water -  The geodes that contain water are referred to as "enhydros" geodes.  The enhydros geodes are frequently found either in streambeds (where a crack developed in the geode and water seeped in), or more often they are encountered when the geode is dug "in situ" (in place from its host rock).  Geode shells are porous and will take on water over time especially where their environment may not be disturbed by outside forces to a great extent.  Water will find its way into an open space eventually.  Some locations contain a high percentage of enhydros geodes while others contain none and this is partially explained by the characterization of the geode shell and local geological conditions.  Shells that are highly developed or "well formed" are those that frequently contain water.  There is a degree of speculation as to whether or not the water in some geodes is of very old or recent origin - research by Sinotte (1969) indicated that the pH of the water varied considerably from sample to sample.  Additionally, some enhydros geodes tend to dry up after being removed from their host rock, with some samples drying up a few days to several years after collecting.  While uncertain and still under study, some geodes probably contain very old water and some very recent, but just how old the water may be from sample to sample is difficult to determine.  Some geodes just have a few drops while some larger geodes have contained enough water to partially fill up a gallon jug. 

Oil (bitumen) -  The geodes that contain oil (bitumen) are found in an area near Niota in Hancock County, Illinois.  Research indicates that nearby oil deposits from Pennsylvanian-age rocks may somehow be linked to these interesting geodes.  Oil is able to migrate quite readily through certain soils and rock, and the conditions at the oil geode location may indicate a favorable environment for oil to reach the surface as the host rock that contains the geodes at this location is also frequently covered with a slick of oil.  NOTE OF CAUTION TO COLLECTORS:  These geodes, while interesting, are very messy to not only mine but also to have or store in a collection.  Make sure to store any collected geodes in a location that is free of dust and also that has a bottom that will not leak and can get dirty.  The oil has a way of soaking through anything permeable over time.

Figures 19 a, b:  Quartz (Chalcedony) lined Keokuk geode with oil (bitumen).  In the bottom portion of the geode (half on the right), enough oil accumulated and formed a flat plate-like structure on the chalcedony.  The geode is approximately 5 inches across.

Figures 20 a, b, c:  Quartz (Chalcedony) lined Keokuk geode with a large quantity of oil (bitumen) for the size of the geode.  When the geode was opened, the oil was thin enough that it was leaking out so a mason jar was quickly sought and the oil collected and saved from this particular sample.  The geode is approximately 4 inches across.


Minerals Primarily Found in the Keokuk Formation near Keokuk, Iowa

Violarite- This mineral may be the cause of black phantoms in calcites, notably at Sheffler's Mine (positively unidentified).  Research is ongoing, especially since violarite is a relative to millerite, presumably not found in Warsaw Formation geodes.  Highly suspect given lack of nickel related minerals in the Warsaw Formation specifically.  Specimens reported likely contained pyrite as inclusions.

Polydymite- Another relative to millerite - this has been confirmed in geodes from the Keokuk Formation at scattered sites in the Keokuk area as a weathering product of millerite.  The author has found specimens of polydymite in association with millerite and malachite in chalcedony-bearing geodes from the Upper Keokuk Formation in the vicinity of Hamilton, Illinois.  Not known in the Warsaw Formation.

Millerite- Millerite occurs in geodes/vugs located in the Keokuk Formation. Most often the rods or needles you will find in geodes from the Keokuk area (Warsaw Formation) are capillary pyrite or marcasite (see above), not millerite (credit for Keokuk Formation to Sinotte, 1969), but research continues.  If you think you have a Warsaw Formation geode that contains millerite, please report it!  This would be a significant finding for Keokuk geodes from the Warsaw Formation!


Credits and Other Notes on Keokuk Geodes

Credits for this page are extended to Dr. David Hess, Professor Emeritus at Western Illinois University, Dennis Haas, Steve Sinotte (1969), Mark Sherwood, and others who have contributed to the recent research on Keokuk geodes and who have helped in the crystallography of the Keokuk geode minerals listed on this page.  The mineralogy of Keokuk geodes is an exciting and fairly well-focused area of research that the author of this page has and continues to enjoy, and with the help of other collectors and geologists in the area, perhaps we can work together to shed more light on the formation and mineralogy/crystallography of Keokuk Geodes.

The list above is one of the longest lists of minerals found in geodes in the world, other than the Indiana geodes which do have some of the same minerals and other minerals as well, and the Chihuahua "coconut" geodes from Mexico (presumably formed differently from the Midwestern geodes but also with an extensive mineral list). This is one reason why Keokuk geodes are so sought after and why rockhounds are frequently found in the Keokuk, Iowa area. There is not a weekend of the year when the weather permits is there not a rock collector out somewhere in the Keokuk area hunting for these objects of curiosity and beauty.  Unlike other rocks and minerals, every single geode is different so you never know what you may find once opened!

Unless someone comes forward with samples, the following minerals have been reported and published as being in Keokuk geodes but (to date) have not been confirmed:  Fluorite (mistaken for very well-formed, usually iridescent cubic brown calcite - a cleavage test will confirm calcite and not fluorite.  Calcite has three directions of cleavage while fluorite has four), Galena (mistaken for either sphalerite or large modified pyrite crystals), Ankerite (possibility, but Sinotte (1969) reported all possible ankerite as being dolomite given iron quantities present), Celestite (mistaken for barite) - celestite is found in Indiana geodes in locally abundant quantities, Magnetite (old reference - most likely goethite but sample analyzed unknown).  Additional minerals most likely exist in Keokuk geodes, but require additional research.  If there is something really "odd" in your geode that you don't know what it is, it would be interesting to have the geode analyzed further to see what exactly it is you have.

The majority of Keokuk, Iowa area "red quartz geodes" that have been sold in the past several years are fakes - they are simply yellow-orange iron-stained geodes that have been put in an oven and heated to a certain temperature where the yellow-orange iron turns to red.  While these geodes are different, they are not naturally occurring and is another example of a marketing tactic to make more money (see paragraph below), similar to what is done to lower grade Brazilian amethyst geodes that are heated to turn the crystals yellow-orange, and then sold as "citrine geodes".  Buyer beware!  As the old saying goes, if it is too good to be true, it probably is.  Most natural red quartz geodes exhibit a red hematite staining that will easily wash off the surface of the quartz, leaving only a hint of the pink or red coloring behind.

Some Keokuk geodes (from both the Keokuk and Warsaw Formations) also resemble fossils (see photos below). 

Figures 21 a, b, c:  Keokuk geode (from the Warsaw Formation) resembling a horn coral fossil.  Note the perfectly preserved horn coral "tip" in the middle photograph.  This particular geode was collected from the Lower Warsaw Formation at Sheffler's Geode Mine (old mine).  The interior is lined with quartz and has several calcites as well (right photograph).  Specimen measures approximately 3 inches in diameter.

Keokuk geodes range anywhere from the size of a garden pea up to large pumpkins (30" across). Large, hollow geodes above 2 feet in diameter are still being found.  The average size of the typical Keokuk geode is between 2-5 inches, but this varies greatly from one geode-bearing exposure to the next (in our experience, some exposures have contained geodes only up to 4 inches in diameter while others have contained geodes that were primarily 5-6 inches in diameter or larger, and few smaller!).  Many locations also contain geodes of all sizes from peas to pumpkins!  It depends on the layer of the Lower Warsaw Formation you are collecting in and the location of the exposure in most cases.  In our experience, the best mineralization is found in the average sized or slightly larger geode, anywhere from 2-7 inches.  There is no doubt that some fantastic specimens a foot or more in diameter containing excellent secondary mineralization exist and are still being collected, but in terms of volume, and in our experience, the geodes in the 2-7 inch range contain the highest quality secondary and/or greater mineralization.

The percentage of hollow geodes dug "in situ" (from the Lower Warsaw host rock) varies not only from location to location but also in the same outcropping.  The percentage of hollow geodes in some outcrops can be as little as 5-10% or as great as 80-90%.  In large exposures where numerous geodes are present, certain areas in the outcropping contain different mineralization and/or a different percentage of hollow geodes compared to an area in the outcrop several feet away. 

I do not feel that any mineral or mineral combination in Keokuk geodes can be marked or labeled as being "rare". I have been collecting Keokuk geodes for the past 18 years, and with the cooperation of several landowners in the area, all 18 minerals positively identified in Keokuk geodes with the possibility of new minerals not previously known to exist and currently under study were found in just a matter of a few years with upgrading to previous examples now taking place. Of course, some minerals such as malachite in Keokuk Geodes are presently much less frequent than your basic quartz geode, but are still not considered to be "rare" in my opinion as this mineral in addition to other minerals can be found in some exposures quite readily. With enough knowledge of the Keokuk area and extremely localized geology in relation to elevation and make-up of the strata (and a little bit of luck, time, and patience), any collector can find an exposure of geodes with the desired, or needed, mineralization with at least a little effort.  The "rare" term has been used especially as a marketing tactic (similar to other collectibles, but a finite number of many other collectibles is known, such as many coins or toys, for example) in recent years when selling a Keokuk geode and has led to confusion and misleading information among especially newer or amateur Keokuk geode collectors.  There are a few ranked lists on other websites that contain Keokuk geode interiors or minerals listed from rarest to most common - these lists have no relevance and are largely a function of that collector's knowledge of locations and collecting experience, where they have access to collect compared to other collectors, or simply a way to hopefully make more money on what is listed as the "rarest" as these lists can and do change over time.  I have received comments concerning this paragraph with some agreeing and others not agreeing with my statements.  Looking at this objectively, there is NO KNOWN FINITE NUMBER of specimens containing a certain Keokuk geode mineral so in addition to previous reasons listed above, I do not feel a "rare" label can be assigned to certain Keokuk geode minerals and/or types.  What may appear to be less frequent or locally abundant now may become very frequent and abundant in the future with the collection of additional specimens or the expansion/opening of a new or present location.  "Rare" in relation to Keokuk geodes is a term of subjectivity (definitions vary from person to person) and should never be included in any scientifically-related endeavor where objectivity is more widely accepted and the better method to proceed upon.

Page composed by Mike Sandstrom, owner of Geode Gallery, LLC.  I welcome any additions or questions you may have related to the Keokuk geode information presented here.  Please contact me at .  I will post additional findings to this page when appropriate.  If you have photos of Keokuk geode minerals I would also be interested in seeing them (they won't be posted on this page unless you provide permission and it isn't a duplicate of something already posted above - PHOTO SUBMISSIONS:  If you intend on submitting photos, PLEASE compress or reduce individual image sizes to preferably less than 500KB in size for quicker downloading and disk space conservation).  I hope in the future to post additional pictures of geode minerals and photos of geodes in their host matrix or "in situ" at various locations when time allows to help the collector visually identify geodes in the field when collecting and to assist in proper identification of minerals in Keokuk geodes.

Please check back in the future for more updates.  All photos and text presented on this page are the property of the author (or other individuals where cited) and should not be used without permission from myself or the owner of the photo.  Content may be used in documents or articles with proper citation but any photos or text used in the content of formally-published articles, promotions, or presentations must be accompanied by written permission from the author and cited fully.  Copyright - Geode Gallery, LLC (2014).  Page last updated January 14, 2014.


Hess, D.F. and M.A. Sandstrom, 1998.  Geodes:  Origin and Mineral Inclusions.  Formally un-published in a refereed journal but locally peer-reviewed for acceptance as a handout - distributed as a handout at Betty Sheffler Day, Sheffler's Geode Mine, September 1998. 8 pp.

Holden, Martin, 1999.  The Encyclopedia of Gemstones and Minerals.  Michael Friedman Publishing Group, Inc., 303 pp.

Sinotte, S.R., 1969. The Fabulous Keokuk Geodes, Volume I.  Wallace-Homestead Company, Des Moines, Iowa, 292 pp. (Out of Print).

Return To The Geode Gallery Homepage