Aowa Creek Specimens

masthead saying 'We Proceeded On'

Reprinted from We Proceeded On1

Greenhorn Limestone

Ponca State Park

Small cliff of light-colored, banded rock

Photo by the author

The formation beneath the prominent cliff face of Greenhorn Limestone here in Ponca State Park is the gently sloping Graneros Shale, which contains layers of the mineral melanterite. Melanterite, formed after the decomposition of pyrite or marcasite, can have a striking bluish green color that the captains either mistook for cobalt or correctly identified as "copperas," and it's a strong candidate for one of the tested minerals that debilitated Meriwether Lewis on August 22, 1804.

Worthy of Experimentation

In the early morning of August 22, 1804, with the memory of burying their "Decesed brother"2 Sergeant Charles Floyd still fresh in their minds, Meriwether Lewis, William Clark, and the Corps of Volunteers for North Western Discovery pushed off from their camp for another day's struggle against the unrelenting current and stiff winds along the Missouri River. It already had the makings of an important day as the captains had decided the time had come for a vote to determine Sgt. Floyd's replacement. But the focus of this day changed some two and a half miles out of camp.

While laying up at present-day Aowa Creek, Nebraska to meet two hunters who had brought in game, a distinctive feature of the landscape attracted the captains' attention. Either from their own observations, or from a sample brought in by the inquisitive John Shields, the captains noted that the bluffs on the larboard side of the Missouri River possessed something more interesting than any of the rocks they had encountered since embarking from St. Louis.

And so it was that Meriwether Lewis probably retrieved a copy of Richard Kirwan's Elements of Mineralogy from a keelboat locker to re-acquaint himself with the procedures of assaying and identifying mineral deposits. Here, at the mouth of Aowa Creek, were rock exposures that were worthy of experiment.

Lewis was already aware, by virtue of his scientific training in Philadelphia in the spring of 1803, that many of the diagnostic mineral tests in Elements of Mineralogy required smelling the fumes emitted from specimens. This was true whether the sample in question was arsenic ("this Calx is often found in Clays and blue Marles and may be distinguished by its smell when thrown on burning Coals"3) or other ores ("In the dry way, it is discovered by its evaporation, in the form of a white Smoke on burning Coals or red hot Iron, with its peculiar smell").4

Viewing what he and Clark thought might be cobalt, Lewis knew that "When Arsenical it [cobalt] gives out that smell on burning coals."5 So Lewis headed up to the bluffs to pursue his scientific duties, undoubtedly pleased that he could begin to fulfill President Thomas Jefferson's objective of ascertaining "the mineral productions of every kind" in this new terrain.6

Calcerous Concretions in Carlile Shale

West of Vermillion Ferry

An man stands on a white, steep hill overlooking the Missouri River before dams


While selenite crystals may be observed throughout the Graneros, Carlile, Niobrary, and Pierre, they are most abundant on slopes of the Carlile formation, especially at the Ionia Volcano or on other bluffs in that region, where they are commonly mistaken for mica. They vary in size from minute needles to 2 to 3 inches across. They originate near the surface from the weathering of the iron pyrites scattered through the clay, some of the products of which react with lime in solution in percolating waters, forming calcium sulphate which crystallizes as selenite. The usual belief that they were formed when the strata were deposited and that they occur in large numbers throughout the beds is therefore erroneous.

Source: George Evert Condra, Geology and Water Resources of a Portion of the Missouri River Valley in North-eastern Nebraska (Washington, D.C.: Government Printing Office, 1908), 12–13.

—Kristopher Townsend, ed.

Selenite Crystal

Long, clear and colorless crystal

Photo by Eurico Zimbres. Permission under the Creative Commons Attribution-Share Alike 2.5 Generic license.

Selenite Crystals on Greenhorn Limestone

Crusty cliff face

Photo by the author

Selenite crystals cover entire fracture surfaces of the Greenhorn Limestone in a locale passed by the expedition on August 22, 1804. Selenite, a transparent, colorless variety of gypsum, along with pyrite, marcasite, and melanterite, were among the minerals that William Clark observed in "great quantity" along this route that would comprise many of the collected specimens that Lewis shipped back East from Fort Mandan in April 1805.

Aowa Creek Exposures

Clark vividly described the rock exposures at the mouth of present-day Aowa Creek,7 just a few miles upriver from their encampment on August 21, 1804.

Commencement of a Bluff on the L.S . . . . This Bluff contain Pyrites alum, Copperass & a Kind Markesites. also a clear Soft Substance . . . . Capt lewis was near being Poisened by the Smell in pounding this Substance I belv to be arsenic or Cabalt.8

examonation of this (1) Bluff Contained alum, Copperas, Cobalt, Pyrites; a alum rock Soft & Sand Stone. Capt. Lewis in proveing the quality of those minerals was near poisoning himself by the fumes & tast of the Cabalt which had the appearance of Soft Isoglass- Copperas & alum is verry pure.9

The bluffs at this location are primarily comprised of Cretaceous age Graneros Shale and Greenhorn Limestone.10 Clark did well here to mark the presence, probably in the Graneros Shale, of pyrite and marcasite, both iron sulfide minerals with an identical chemistry but different crystal forms.

Sgt. John Ordway provided helpful detail as well when he described one of the minerals as "brass," which is the signature color of pyrite and marcasite, and recorded that the minerals had a "Sulpheras Smell,"' another positive indicator of iron sulfide minerals.11

An Experiment Goes Badly

Mineral specimens were collected here, probably by Lewis himself. For at least one specimen, Clark (in the Field Notes) described Lewis as "pounding this Substance"12 and he recorded in his Notebook Journal that Lewis went further, inhaling the fumes and tasting the specimen(s). The way both the field notes and journal entries are presented (presumably chronologically), it could be surmised that Lewis performed the tests at the site of the collection, because Clark described Lewis's physical condition right after his description of the mineral content of the locale, followed by the subsequent events of the day.13

The experiment went badly. While there is uncertainty regarding what minerals Lewis attempted to assay using this method, the fumes and ingestion of these substances made him ill, with an aftereffect that would linger for several days. It's probable that Lewis hoped he could recover on his own, but Clark reported that his friend took "a Dost of Salts this evening to carry off the effects."14

Was it Copperas?

Farther upriver in what is present-day Ponca State Park, one of the most complete and accessible outcrops in this region lies adjacent to the park's boat ramp. This exposure is approximately 5.5 miles upriver from the bluff that Clark described earlier on August 22nd, so it probably is not the exact spot that Clark noted as being "Seven miles above."15 Nevertheless, the river-facing bluffs along this route expose two of the same geological formations the captains encountered at the mouth of Aowa Creek:

Clift of Allom Stone of a Dark Brown Colr. Containing also in crusted in the Crevices & Shelves of the rock great qts. of Cabalt, Semented Shels & a red earth.16

The formation at ground level is the gently sloping Graneros Shale, a medium to dark gray, partly calcareous shale with interbeds of bentonite, siltstone and sandstone, which is overlain by a prominent outcrop of Greenhorn Limestone, a light gray, ledge-forming limestone with interbeds of marl and calcareous shale.17

To check on Clark's journal description, it was critical to see if any incrustations could be found in the crevices of the rock or in association with 'Semented Shels'. During one visit, I picked a route between fallen slump blocks to try to reach the sharply defined ledges near the top of the bluff. While passing upward through the Graneros Shale, I observed rock fragments covered with the iron sulfide mineral marcasite.

Although it was not evident to me on that day, melanterite,18 a hydrated iron sulfate (FeSO4) formed after the decomposition of pyrite or marcasite (iron sulfide or FeS2) is also present in the upper section of the Graneros Shale. Melanterite can have a striking bluish green color that the captains either mistook for cobalt or correctly identified using the antiquated term "copperas."19

Copperas (which will be referred to as melanterite hereafter) is a strong candidate for the mineral that Lewis tested on August 22nd. When this mineral is formed from the oxidation of iron sulfide, the reaction also produces sulfuric acid (H2SO4), a highly corrosive substance that can seriously damage the skin upon contact or the lungs when inhaled. This suggests that Lewis may not have needed to "roast" this mineral because just "pounding" the specimen and coming into contact with the associated sulfuric acid would have deleterious effects.20

There's one piece of compelling internal evidence for this theory in addition to the fact that Clark specifically identified "copperas" as being present at the outcrops they were examining. Sometime after working through the potential chemistry of what Lewis and Clark were encountering at these outcrop exposures, I updated the antiquated descriptive terminology in the explanatory notes of the mineral specimens that Lewis sent back East from Fort Mandan.21 Shortly after the Fort Mandan mineralogical specimens arrived in Philadelphia, Adam Seybert, physician, gentleman scientist, and leading mineralogy expert, added supplemental mineralogical identifications to augment Lewis's original descriptions.22

An extensive annotation Seybert added to Lewis's description of Fort Mandan mineralogical specimen No. 4, collected on August 23, 1804, observed that it "consists principally of Sulphat of Iron derived from decomposed Sulphuret of Iron."23 Remarkably, Seybert described the same reaction of iron sulfide (Seybert's "Sulphuret of Iron") to form melanterite (iron sulfate or Seybert's "Sulphat of Iron") that occurs in nature within the bluffs of northeastern Nebraska, thus providing additional evidence that this mineral was present in these localities for Lewis and Clark to encounter.

On my investigation of the bluffs, I finally reached the basal unit of a superb exposure of the Greenhorn Limestone.24 It is deceptive in its initial appearance because it seems to be comprised of different colored rocks, grading from a light gray in some sections to a color resembling hazelnut coffee creamer in others. However, it is all the same formation. Like most limestones, including those Lewis and Clark examined at the Three Forks of the Missouri, this formation is a gray color in fresh exposures, but weathers to a light tan to cream color once it has been exposed for a period of time.

In examining some of the Greenhorn Limestone slump blocks downslope, I was soon rewarded with a spectacular view of selenite crystals (a transparent, colorless variety of gypsum) that had formed in the fractures in the limestone. These selenite crystals, a hydrated calcium sulfate (CaSO4-2H2O) are striking enough in appearance to justify the captains' interest in them, and were second only to the number of iron sulfide minerals, and their reaction byproducts, that they observed and collected.25

Selenite covers broad surfaces of these rock fractures, forming when sulfuric acid reacts with the calcium carbonate (CaCO3) of the limestone.26 Also present along bedding planes of the limestone are dense remnants of the extinct bivalve Mytiloides sp. or Inoceramus sp., the "Semented Shels" that Clark notes in his journal.27

Twinned Crystal Selenite

Two clear crystals joining to make an 'X' shape

Photo by the author

Three documented mineral specimens from the expedition that have survived to the present-day have been identified as selenite, including a twinned crystal (Fort Mandan mineralogical specimen No. 8 collected on August 23, 1804) that is similar to this specimen observed in the Carlile Shale outcrop at the Ionia Volcano locale, suggesting that the Carlile Shale was the source of some of the selenite samples that the captains collected in this region.

A Productive Day

To return to the sequence of events on August 22, 1804, Lewis and Clark examined the bluffs at the confluence of Aowa Creek when they landed to pick up two deer that their hunters had killed. Intrigued by what they saw there, Lewis decided to perform his ill-fated experiments, most probably on iron sulfide minerals (pyrite or marcasite) or on the secondary mineral melanterite. Not discouraged by the aftereffects his friend suffered, Clark kept an eye out for additional exposures upriver.

At certain loops of the river, bluffs were right along the shoreline and rock exposures would have been obvious no matter where the keelboat and pirogues were. If Lewis did, in fact, become ill from exposure to sulfuric acid after testing the minerals near Aowa Creek, it would be logical to assume that any further collecting or observation that occurred upriver was conducted instead by Clark. Regardless of who did the work, the result was the single most productive day of collection on the 1804 leg of the expedition. At least nine mineral specimens were collected. On the following day, at least six more samples were obtained, including one of the three documented surviving mineral specimens, Fort Mandan mineralogical specimen No. 8, which has now been identified as selenite.28

The list of mineral specimens collected on August 22nd doesn't indicate at which locality the nine samples were collected, but it can be deduced that some were collected at Aowa Creek. That certainly includes Fort Mandan mineralogical specimen No. 68, recorded as having been "brought [to] us by one of our hunters, John Shields who found it at the Allum Bluff,"29 the same location that Clark described as having "great quantities of those minerals."30

  • 1. John W. Jengo, "'Blue Earth,' 'Clift of White' and 'Burning Bluffs'," We Proceeded On, Volume 37, No. 1 (February 2011), the quarterly journal of the Lewis and Clark Trail Heritage Foundation. Editorial additions include page titles, side headings, and graphics to assist the web-based reader. The original printed format is provided at
  • 2. Gary E. Moulton, ed., The Journals of the Lewis and Clark Expedition, (Lincoln: University of Nebraska Press, 1986), 2:495. All Lewis's or Clark's journal quotations between August 22-24, 1804 in the ensuing text are from Moulton, Journals, Vol. 2, by date.
  • 3. Richard Kirwan, Elements of Mineralogy, Salts, Inflammables, and Metallic Substances (London: Printed for P. Elmsly, in the Strand, 1796), 2:259-260.
  • 4. Ibid., 438.
  • 5. Ibid., 273.
  • 6. For an in-depth discussion of the mineralogical objectives of the expedition and the role of Thomas Jefferson in this enterprise, see John W. Jengo, ''Mineral Productions of Every Kind: Geological Observations in the Lewis and Clark Journals and the Role of Thomas Jefferson and the American Philosophical Society in the Geological Mentoring of Meriwether Lewis," in Robert S. Cox, ed., "The Shortest and Most Convenient Route: Lewis and Clark in Context," Transactions of the American Philosophical Society, Vol. 94 (2004), Part 5, 136-214.
  • 7. Aowa Creek has been correlated with Clark's dream-inspired name of "Roloje" Creek; Clark described this stream as coming "in from the L.S. passing under the Clifts for several Miles," meaning the creek was passing under bluffs that bordered the Aowa Creek drainage, and not those cliffs that face the Missouri River. Perhaps Clark got this information from Shields or the other unnamed hunter they were meeting at this confluence, or he explored this creek himself while Lewis was collecting and testing minerals.
  • 8. Clark, August 22, 1804, Moulton, Journals, 2:500. Also referred to as Clark's "River Journal," these Field Noteswere first-draft notes presumably recorded under less favorable weather conditions that were later copied into the Notebook Journals.
  • 9. Clark, August 22, 1804, Notebook Journal in Moulton, Journals, 2:501. These entries were recorded in the renowned small, stenographer-like notebooks that the captains intended to be the official journals of the expedition.
  • 10. Raymond R. Burchett, Vincent H. Dreeszen, Vernon L. Souders, and G.E. Prichard, Bedrock Geologic Map Showing Configuration of the Bedrock Surface in the Nebraska Part of the Sioux City 1° x 2° Quadrangle, (U.S. Geological Survey Miscellaneous Investigations Map I-1879, 1988), one sheet, scale 1:250,000. The Cretaceous Period covers the time period from approximately 145.5 to 65.5 million years ago.
  • 11. Moulton, Journals, 9:42.
  • 12. Moulton, Journals, 2:500.
  • 13. There appears to be an error on Martin Plamondon's cartographic reconstruction of this day's travel; see Martin Plamondon II, Lewis and Clark Trail Maps: A Cartographic Reconstruction, (Pullman: Washington State University Press, 2000), 1:105. I have taken Clark's initial S47°W and subsequent due West course and distance call from the previous day's encampment, and ended up on the south side of Aowa Creek whereas Plamondon took these bearings and ended up over a mile away on the north side of the creek. By my reckoning, it appears that Plamondon mistakenly used a S70°W bearing instead of S47°W. This is very important because it determines the location of the mineral bluffs that Lewis sampled and Clark described. The reason I may be correct is that Clark had Aowa Creek coming in "above this Bluff," which would place the mineral bluffs south of the creek's confluence with the Missouri River. In ground-truthing this supposition, I found numerous exposed bluffs along 594 Avenue on the south side of present-day Aowa Creek that are promising candidates for the bluffs Lewis initially sampled, but there are also fine bluff exposures north of the creek that the captains may also have examined.
  • 14. Moulton, Journals, 2:500.
  • 15. Ibid, 501. There's an internal discrepancy associated with this distance call. Clark's course and distance for this transect adds up to eight miles, not seven, and Clark appears to have replaced "eight" miles with "seven" miles in his notebook journal entry for some reason. Another issue about where this distance call ends up (which determines which outcrop the captains observed later in the day) relates to where the distance call starts (north or south of Aowa Creek), which is affected by the bearing error described above.
  • 16. August 22, 1804, Moulton, Journals, 2:501.
  • 17. Burchett, et al., Bedrock Geologic Map, USGS Map I-1879.
  • 18. Roger K. Pabian and Dennis R. Lawton, Geology of Ponca State Park, Nebraska, Educational Circular 6 (Lincoln: Institute of Agriculture and Natural Resources, 1984), 32.
  • 19. The blue color is derived from impurities of copper that replace some of the iron in the crystals; the name "copperas" is from the Greek meaning "copper water."
  • 20. Contact with sulfuric acid will burn the skin, and breathing sulfuric acid causes respiratory tract irritation. Melanterite is one of the few sulfates that are soluble in water, and that may have been the manner in which Lewis ingested it; if so, in combination with sulfuric acid, Lewis could have experienced burns to his mouth, throat, and stomach, triggering gastrointestinal distress.
  • 21. The Fort Mandan mineralogical specimen numbers used in this article follow those that were recorded in the Donation Book of the APS, see Gary E. Moulton, ed., The journals of the Lewis and Clark Expedition, August 25, 1804-April 6, 1805 (Lincoln: University of Nebraska Press, 1987), 3:473-478. Any reference to a mineral specimen in the narrative prefaced by "Fort Mandan mineralogical specimen" refers to those minerals primarily collected in 1804 and sent back East from Fort Mandan in April 1805. All Lewis or Clark journal quotations between August 25-September 1, 1804 in the ensuing text are from Moulton, Journals, Vol. 3, by date.
  • 22. For details on the important role that Adam Seybert played in the subsequent fate of the entire mineral collection, see John W. Jengo, "'Specimine of the Stone': The Fate of Lewis and Clark's Mineralogical Specimens," We Proceeded On, Vol. 31, No. 3 (August 2005), 17-26 (also on this site as Mineralogical Scientists).
  • 23. Moulton, Journals, 3:473. The term "sulpheret" was in common use in this time period and was subsequently replaced by the term "sulfide." This transition was occurring around the time of the publication of William Barstow, Sulphurets: What They Are (New York: A. Roman and Company, 1867), 114.
  • 24. Robert F. Diffendal, Jr. and Anne P. Diffendal, Lewis and Clark and the Geology of the Great Plains, Education Circular No. 17 (Lincoln: Conservation and Survey Division/School of Natural Resources, December, 2003), 27.
  • 25. The nine samples collected on August 22, 1804 are Fort Mandan mineralogical specimens Nos. 10, 13, 18, 20, 38, 49, 51, 56, and 68; most of these appear to have been iron sulfide minerals (e.g., pyrites) and so-called crystallized "Sulphat of Lime," or selenite. Lewis was apparently collecting different selenite crystal configurations not knowing they were chemically the same mineral. Appreciative thanks to Sandy Schenck of the Delaware Geological Survey and Dr. Peter Leavens of the University of Delaware for confirming the mineralogy of the Greenhorn Limestone selenite samples collected by the author.
  • 26. The fascinating relationship between the iron sulfide minerals pyrite and marcasite oxidizing to form melanterite and sulfuric acid, with the sulfuric acid reacting with calcium carbonate to form selenite and reacting with aluminous and potassic-containing rocks to form a potassium aluminum sulfate (alum), intimately links all their parent geological units into one enormous chemical laboratory.
  • 27. Moulton, Journals, 2:501. In the geological literature, these fossils are identified as Inoceramus, which used to be synonymous with genus Mytiloides, but recent paleontological research has suggested that these extinct bivalves should be distinguished from one another. Given the regulation prohibiting collection of fossils within the park, I was unable to acquire samples for further study to determine which genus identification (Inoceramus or Mytiloides) is correct. The differences are very subtle and require detailed morphological analyses.
  • 28. Jengo, "Specimine of the Stone," 23.
  • 29. Moulton, Journals, 3:478.
  • 30. Moulton, Journals, 2:500.