THE ORIGIN OF BANDED AGATES
Robert G. Welch
Part Two
It is my belief that all banded agates are formed by a process called diagenesis. Diagenesis is a
geologic term meaning changed from its original form after deposition. For example, a basaltic lava
flow cools after deposition and as it cools, it shrinks. This causes shrinkage cracks, much like the
ones formed in drying mud. This is the first stage of diagenesis. Rain falls on the lava dissolving
minerals from it and deposits new minerals as cement, coatings or crystals in the cracks and voids,
thus altering the composition of the lava. This is diagenesis resulting from sub aerial exposure. If the
lava lays exposed for a period of time with nothing being deposited on top of it, the minerals formed
will usually be deposits like calcite, aragonite, zeolite, hematite or limonite formed by the weathering of
minerals in the lava bed itself. I have in my collection a hematite "plume" from the Woodward Ranch
in Brewster County, Texas that was present in an otherwise empty void. If agate had later been
deposited around it, it would have been a red plume agate.
Lava flows may be exposed for many years before volcanic ash or another lava flow is deposited on
top of it. If another flow is deposited, then this flow will be barren of agate. If the next event is
volcanic ash, then the scene is set for the formation of agate. When rain falls on the ash, the minerals
are weathered, releasing free silica as colloids. As each rain soaks through the ash it picks up
increasing amounts of silica, which drains into the cracks in the lava. When the cracks intersect a
vesicle, the fluid slowly spreads into it, held to the walls by surface tension. If too much comes in at
one time, it puddles in the bottom of the vesicle, forming flat bands.
Each vesicle in the lava is a separate diagenetic environment. What happens to the vesicle is related
to many factors. If the vesicle is not intercepted by a fracture, it may remain empty. The size and
orientation of the intersecting fracture can effect the amount and pattern of filling. Vesicles with no exit
fractures would be prone to flat banding. Small fractures or fractures not at vertical angle could cause
less material to come into the voids. Position in the lava could effect whether small rain events
reached the void. Separate voids can have variations in the chemical environment resulting in a
different colors in adjoining agates. Variation within the agate can be due to differing rainfall, changing
vegetation above, renewed ash falls and other factors.
From the Jurassic to the present extensive volcanism was occurring in what is now the western United
States and Mexico. Many of these events culminated in the ejection of huge amounts of volcanic ash
that were carried by winds across the entire continent. Ash from the two most recent major events in
Wyoming and New Mexico can still be found buried in valleys along some streams. Geochemistry
and radiometric dating can identify both the age and source of these ash falls. The White River
Formation of South Dakota that preserved a large fauna of Oligocene land animals was formed from a
similar event. The vagaries of wind deposited large amounts of ash in some localities and little or none
in other nearby localities.
Agates have been found in limestones across the country from Montana and Wyoming to Kentucky,
Tennessee, West Virginia and states in between. They are found in limestones ranging in age from
Pennsylvanian to Ordovician. These deposits all have two things in common: (1) They have been
weathered and voids formed by solution and (2) the agates only occur in the weathered rocks near
the present surface. Pabian and Zarins refer to these agates as sedimentary agates and postulate that
they were deposited with or shortly after the limestone from silica gels formed from siliceous fossils or
volcanic ash.
A good example of this type of agate are the Teepee Canyon and Fairhills Agates found in place in
the Minnelusa Formation of Pennsylvanian age. The Pennsylvanian was a time of many sea level
changes, causing the shallow shelf to be alternately inundated and exposed. The Minnelusa Formation
was deposited in a shallow shelf environment, which alternated between very shallow sea and "sabka"
conditions. A sabka is a relatively flat lying area that is inundated at high tide but is land at low tide.
In the nearby Powder River Basin the Minnelusa produces oil from Aeolian sand dunes, which often
grade into anhydrite (CaSO4). Interbedded limestones also contain vugs filled with anhydrite. The
vugs developed during times of exposure in wet periods with the anhydrite being initially deposited in
the vugs as gypsum (CaSO4·H2O) and then converted to anhydrite after burial.
During the Tertiary Period (65 million years to present) the Black Hills were uplifted into a large
dome. This folding fractured and exposed the rocks to weathering and erosion. As weathering
progressed the fractures in the Minnelusa Formation were enlarged by solution and the anhydrite
converted to gypsum and was dissolved out, leaving small to intermediate size cavities. This system
differs in important ways to from the volcanic system. The limestones are initially more porous than
lava and they can be dissolved by rainfall, allowing infiltration into the host rock itself as well as along
the fractures. There is also no base to the infiltration as is the case in individual lava beds. In
limestone the water passes on through and is drained to springs and caves. This is the reason that flat
(geopedal) banding is absent in agates formed in limestone.
During the time of exposure and erosion, extensive volcanism was occurring nearby in what was to
become Yellowstone National Park and to the west. This volcanism periodically dumped thick
blankets of volcanic ash on the Black Hills. Many altered ash beds occur in the Tertiary sediments
surrounding the Black Hills. As the ash weathered, silica seeped slowly down into the voids. At first
it soaked into the limestone walls of the cavities replacing the limestone and altering it to chert. This
replacement occurred as convex fronts spreading out from the center to form round to oval nodules.
As the walls became less porous the silica began to be deposited as spherulitic crystal bands typical
of banded agates. Smaller voids with small openings into them filled completely. Changes in color
dominance within the agate are due to changes in the composition of impurities and their
concentration. This would occur naturally as weathering progresses in changing climate conditions
with wet period and dry periods and their resultant changes in vegetation.
Roger Clark in his excellent book "Fairburn Agate Gem of South Dakota" has many excellent
photographs of Fairburn Agate along with Teepee Canyon and Fairhills Agate. Fairburn Agates were
eroded from the Minnelusa Formation and re-deposited in Tertiary sediments, particularly the
Chadron Formation. Of particular interest are illustrations of slabs of a single Teepee Canyon nodule
on pages 47-50. Being a Geologist my first impression of this nodule was that it displays typical karst
structure. Karst structure develops by solution and collapse in limestones related to cave formation.
It appears that after a long period of low rainfall, which deposited red and orange bands, there was a
major rainfall event. This event caused water to surge through the developing agate ripping partially
lithofied fragments from the walls and fracturing of some of the remaining wall structure. The jumble
of irregular elongate pieces leaves the impression of "floating fragments" when the rock is slabbed.
After this disruption a new phase of agate deposition filled the rest of the cavity with white to clear
agate.
Why don't I think these agates were formed in the Pennsylvanian Period when the limestone was
deposited? To begin with, the different agate colors are due to different minerals as illustrated by
Roger Clark on page 62 of his book. If these agates formed from a silica gel deposited with the
limestone, all of the bands should be the same basic color. Instead they contain minerals of different
compositions. Some were even deposited in varying chemical environments.
There is no evidence of volcanism during the Des Moinesian Epoch of the Pennsylvanian Period in this
region. There is therefore no source of silica to form agates. Even creatures using silica in their
bodies do not become abundant enough to be the source unless they have a good source of silica in
their water. A mechanism for dissolving this organic silica and re-depositing it as silica gel would need
to be found before this could be considered as a source
In conclusion, the evidence available fits an interpretation that agates form by a process of diagenesis
in pre-existing rocks. They form above the water table where drying can occur between influxes of
silica. A silica source that will release large amounts of silica at one time is required. This source can
only be finely divided volcanic ash from basic igneous sources.