Saturday, January 25, 2014


Winnemuca Lake petroglyphs. Photograph: Larry Benson.

A recent scientific investigation of petroglyph boulders on the west side of the Winnemucca Lake basin in Nevada has yielded hard dates on the age of the petroglyphs.

Paleoclimatologist Larry Benson (an emeritus scientist with the U.S. Geological Survey who does research for the University of Colorado and its Museum of Natural History) had noticed that the symbols are much whiter than the gray rock they're carved into.

Winnemuca Lake petroglyphs. Photograph: Larry Benson.

Benson needed permission from the Pyramid Lake Paiute Tribe to sample the rock coating. He did finally get permission to sample the coating on rocks near the petroglyphs although he has not yet been allowed to sample any of the ancient rock art.  The whitish coating proved to be Calcium carbonate (CaCO3) and had been deposited when rising lake water lapped over the lower portions of the petroglyph boulder.

Winnemuca Lake petroglyphs. Photograph: Larry Benson.

His paper in the Journal of Archaeological Science, 40 (2013), 4466-76, Dating North America’s oldest petroglyphs, Winnemucca Lake Sub-basin, Nevada, by L.V. Benson, E.M. Hattori, J. Southon, and B. Aleck, outlined the reasoning involved. Benson was aware that Carbonate crust could not have been deposited in the petroglyphs from the lake water unless they were actually underwater at some period. This period was determined using two methods. Laboratory analysis "determined the amount of Calcium carbonate in layers of lake sediment over time. When the amount was close to zero, the lake covered the lower part of the mound below 1206m and the petroglyphs below this level. When the value was relatively large, the lake had fallen below the mound and the petroglyphs and made them accessible for carving." (Benson)

Additionally, they detected a fresh-water plankton (Stephanodiscus hantzschii) found today in lakes in British Columbia in those layers proving that a large quantity of fresh water was injected into the lake water. During those periods water level would have been high and the rocks partly covered, thus the carbonate. They also assumed that the petroglyphs were carved during a period when water was low and people could walk to the site. This is, of course, a very simplified explanation and readers are encouraged to refer to the original paper for the scientific details. The second method of age determination was to 14C date the carbonate layer itself. Fifteen carbonate samples were taken near the petroglyphs and  were 14C dated at the University of California-Irvine W.M. Keck Carbon Cycle Accelerator Mass Spectrometry Laboratory.

Winnemuca Lake petroglyphs. Photograph: Larry Benson.

The results in Benson’s words were “to provide a minimum age for carving of the low-elevation (1202-1206 m) petroglyphs, we dated the carbonate crust that coats the petroglyphs. The six carbonate-crust samples from the petroglyph site (WDL12) exhibited an age range of 10.23-9.77 ka with one outlier at 8.69 ka. As the sample abrasion process did not always reach the inner (oldest) part of the carbonate crust, we conclude that initial deposition of the carbonate crust occurred at 10.2 ka and continued until 9.8 ka, a conjecture consistent with the TIC data discussed in Section 3.5, which indicates that lake level was constrained by overflow at 1207 m until 9.3±0.1 ka. (Benson 2113:4473)

Additionally, the time frames indicated by the sediment coring supported that by indicating “the TIC records resulting from the two age models indicate that the base of the petroglyph site was subaerially exposed between 15.0 and 13.2 ka and was subject to the carving of petroglyphs. However, the TIC records resulting from the two age models indicate different times of possible subaerial exposure after 13.2 ka. One age model (Fig. 5A) indicates that the base of site WDL12 was subaerially exposed between 11.3 and 10.5 ka and the other age model (Fig. 5B) indicates that the base of site WDL12 was subaerially exposed between 11.5 and 11.1 ka.” (Benson 2013:4473) Applying another age model gave Benson an age range of 11.3 – 10.5 ka. (Benson 2013:4476)

By combining the date ranges from sediment coring and 14C testing on the carbonate layer Benson could state “We, therefore, conclude that the petroglyphs were carved sometime between 14.8 and 10.2 ka.” (Benson 2113:4473)

I asked Benson some questions based upon my own observations (and lack of detailed knowledge). First, I could imagine Calcium carbonate molecules floating around in the lake for hundreds or thousands of years until rising water brought them to a position to be deposited upon the petroglyph rocks, “wouldn’t that give an excessively ancient date?” Benson explained that the inrush of fresh water that raised the lake level to cover the base of the petroglyphs also flushed the bulk of the preexisting carbonates out of the lake (remember the fresh water plankton indicating that the brackish water had been greatly diluted and/or displaced. (personal communication). I also asked about the appearance of sharp edges on the lines of some of the carvings. “Did they exhibit any evidence of more recent additions or modifications?” Benson answered that some of the sharp-edged lines actually showed carbonate deposition on their surfaces proving that they had not changed since that event (personal communication). This is seemingly iron-clad, with the results of more than one type of test providing results that agree like this. Indeed, this is significant enough that Archaeology magazine named it one of the Top Ten Discoveries of 2013. (Powell 2014:28)

So, thank you Larry Benson. Bringing the knowledge of different disciplines to work on rock art questions can provide surprising benefits. And while I am at it, thank you Archaeology for including a rock art analysis in your Top Ten list for 2013.

NOTE: I am grateful to Larry Benson for taking the time and effort to correspond with me about this, and for providing a copy of their paper and photographs for my use, and to illustrate this. If any of the technical details above are incorrect it is entirely due to misunderstanding on my part, not any lack of consideration and generosity on Larry's part.


Benson, L. V.
2013    Dating North America’s oldest petroglyphs, Winnemucca Lake, Subbasin, Nevada, Journal of Archaeological Science, 40 (2013) 40 (2013) 4466-76.

Powell, Eric A.
2014    North America’s Oldest Petroglyphs: Winnemucca Lake, Nevada, from Top Ten Discoveries of 2013, Archaeology, January/February 2014, Vol. 67, No. 1, p.28.

Sunday, January 19, 2014


Purgatory Canyon, south of the bear, Bent County,
CO., Photograph: Peter Faris, June 1991.

I had actually thought that I was probably through with Barry Fell but I ran into another couple of examples of Barry Fell’s inaccurate methods. This one can be found detailed in the interesting 1996 private publication by Phillip M. Leonard and William R. McGlone titled A Study of Script-Like Petroglyphs in Southeast Colorado, Mithras Inc., Kamas, Utah. This 70 page booklet outlined their multi-year studies comparing abstract figures or symbols found in southeast Colorado with old-world scripts, especially scripts from the Arabian Peninsula. In this posting I will present the case of one single row of figures found south of the famous bear panel in the Purgatory river canyon.

Leonard and McGlone, Fig. 8, p. 14.

The row of symbols in question is found high above the present ground level due to erosion of the valley bottom (described below). This portion of the cliff face is currently unreachable without a ladder or some other artificial aid. The arrangement of markings on this very interesting cliff can be likened to the stratification of a traditional archaeological excavation because the older markings are high up and were later unreachable. More recent petroglyphs are found lower down on the cliff with some examples of Plains Biographic Style images down just a few feet above the ground. Apparently some early devotee of epigraphic interpretation did go to the trouble to carry a long ladder in because the particular symbols in question have actually been painted in with aluminum paint to make them legible from the ground (personal observation).

Detail, dated trident symbol on right.
Leonard and McGlone, Fig. 8-A, p. 14.

One symbol from this line of characters was dated by Ronald Dorn using cation-ratio dating (this must have been done before the aluminum paint was added but I am not aware of the actual dates of either the testing or the painting). The trident-like character on the right side of subgroup-A dated to 1,975 years plus or minus 200 years BP (before present). (Dorn, McGlone, and Leonard 1990:23-36). It should be noted that Dorn subsequently withdrew the results of all of his dating work citing possible contamination of specimens. I maintained to him that this particular date, given the fact of the stratification on this cliff of rock art by age and height, presented additional evidence that this age may well be accurate.

Detail, dated trident symbol on right.
Photograph: Peter Faris, June 1991.

Leonard and McGlone described it as follows: 
“In the fall of 1981, we visited a well-known petroglyph site in southeast Colorado where there were hundreds of glyphs in an assortment of styles on the base of sandstone cliff faces 80 feet high. Some of the panels are 20 feet or more above the present ground level owing to progressive erosion of the valley floor as evidenced by mineral deposits on the cliff wall. Many of the glyphs are in the Pecked Abstract Style and are so old and heavily patinated they are difficult to see clearly even when the light is favorable. Others, in the Plains Biographic Style, appear to be much more recent, judging from their lack of patination and weathering.

One set of signs in a row (Figure 8A) was published as a “What is it?” in the December 1983 issue of Western Epigraphy with the hope that someone could explain its script-like appearance. Greg de la Castro of Conifer, Colorado, responded, saying he thought they were letters of the Sabaean alphabet. When Barry Fell was informed of this a few months later, he agreed. Although the sequence was short, both correspondents saw the presence of two different pitch-fork-shaped characters as pointing toward Sabaean.” (Leonard and McGlone 1996:13-15)

“On a later trip to the site, we saw that only a portion of the line of characters had been included in the original transcription. A search for Native American styles with similar long sequences of signs was unproductive. Study of the complete set (Figure 8B) and comparison to many alphabets world-wide convinced us that the glyphs corresponded more closely to North Arabian than South Arabian (Sabaean) letters - . When we advised Fell of this  and sent him a better photograph, he sent back a translation using the Safaitic (North Arabian) alphabet. The reading was published in McGlone and Leonard (1986) in order to establish priority of discovery and to stimulate comment." (McGlone and Leonard 1996:15)

This translation was accepted by McGlone and Leonard and published in 1986. It read: "Stayed here to trade, then departed after negotiating an augmented trade agreement - Fasih" (p.202) With the word "Fasih" supposedly representing the signature of either the trader, or the inscriber of the passage. McGlone and Leonard later returned to the site and the story is picked up again in their book from 1996.

"When we returned to the site and carefully recorded the full inscription, we found that the transcription developed by Fell from our photograph omitted two signs, improperly included some from a line below, employed natural rock inclusions as letters, and generally mis-applied the Arabic language in the translation. We asked him not to publish the faulty reading and proceeded to study the regional script-like signs ourselves. Our approach has been to collect groups of the signs and send them to knowledgeable specialists for evaluation and comment.” (McGlone and Leonard 1996:15) 

I have previously criticized Barry Fell for his unscientific method and basing his interpretations on improper evidence and falsified data. In this instance we have published testimony of his errors by two of his (at that time) collaborators. This conclusion is backed up by subsequent statements made to me by Bill McGlone on more than one occasion, that he could no longer abide by Fell’s work due to such errors (purposeful or otherwise) and that he regretted his previous association with Fell (private communication). I pointed out above that this row of symbols has been highlighted with aluminum paint. Bill McGlone always maintained that he and Phil Leonard had nothing to do with that and I have no reason to doubt his veracity. I do not know who applied the aluminum paint to the symbols but it was assuredly done by one of Fell’s collaborators (nobody else has really been interested in those particular symbols) so we can point that back at Barry Fell as well. 

The main thing here is that many of these characters are indeed like characters in Old World alphabets, I do not deny that. This does not mean, however, that this is anything more than a coincidence. I do not accept it as actual writing in any Old World script. McGlone and Leonard were interested in the apparent correspondence between many symbols in southeast Colorado and characters from Old World scripts, but their methods were scientific, and they seldom made claims that they could not substantiate. Fell on the other hand - - - ?


Dorn, Ronald I., William R. McGlone, and Phillip M. Leonard
1990    Age Determination of Petroglyphs in Southeast Colorado, Southwestern Lore, 56(2), 23-36, Colorado Archaeological Society, Denver.

Leonard, Phillip M., and William R. McGlone
1996    A Study of Script-Like Petroglyphs in Southeast Colorado, Mithras Inc., Kamas,               UT.

McGlone, William R., and Phillip M. Leonard
1986    Ancient Celtic America, Panorama West Books, Fresno, CA.

Saturday, January 11, 2014


Of the many symbols found in rock art inscriptions it would indeed be strange if none of them were found in other contexts of Native American art.
Hicklin Springs, 5BN7, Bent County, Colorado.
Photograph: Peter Faris, May 1992.

Panel # 3.B2, Hicklin Springs, 5BN7, Bent County,
Colorado. Drawn by Peter Faris, 25 Sept. 1993.

At site 5BN7(Hicklin Springs) in southeastern Colorado’s Bent County one of the petroglyph panels (Panel# 3.B2) carved in the cliff face shows a grouping of short curved lines - semicircles (horseshoes) in four vertical columns.

My field sketch of the panel allows us to count 26 of these symbols on the panel. This is a relatively common symbol in rock art in southeast Colorado and the west, but this is a particularly good grouping of them. According to Thomas Mails if this symbol is painted upon a horse it represents a horseshoe or horse track and symbolizes a horse taken from an enemy in a horse stealing expedition or a fight. Now my grandfather taught me that when you hung up an actual horseshoe for good luck you hung it this way, with the open side up so the good luck would not run out - the way the semicircles are oriented on the rock art panel at Hicklin Springs. Notice that with the Mails and the Bad Heart Buffalo examples the horseshoe is presented the other way around, with the opening down.

Thomas Mails, 1972, Mystic Warriors of the Plains, Barnes and
Noble Books, New York. Pages 220 (left) and 222 (right).

In his book Mystic Warriors of the Plains (1972), Thomas Mails illustrated a number of such symbols that he identified as being used in horse painting. “Painted exploit symbols used on horses. a, war party leader. b, enemy killed in hand combat. c, owner fought from behind breastworks. d, hail. e, coup marks. f, horse raids or number of horses stolen. g, mourning marks. h, medicine symbol.” (Mails 1972: 220)

Writing in Ledger Book Art: A Key to Understanding Northern Plains Biographic Rock Art, James D. Keyser (1989:92) called Ledger Book Art a “lexicon” for Biographic rock art. In this he was pointing out that Ledger Book art, and Plains Biographic rock art as well, are usually records of specific events and that the symbolism used extends to both media.

Amos Bad-Heart Bull, p. XV, in Wind on the
Buffalo Grass, Leslie Tillet, 1976.

“Amos Bad Heart Buffalo’s drawing of himself as a cowboy, done Dec. 3, 1900. The inscription, translated by Helen Blish reads, “Oglalas from White Clay district herding their cattle.” The sketch at upper left is of a cattle ranch of that time, and the label above it reads, “Chenney River S. Dak. Squn Hamper Creek.” (Tillet 1976: XV) Even though he seems to have reconciled to living the white man’s life, he still has his horse painted with traditional symbols, a holdover of the traditional attitudes and ways.

Amos Bad-Heart Bull, p. 34, in Wind on the
Buffalo Grass, Leslie Tillet, 1976.

Also by Amos Bad Heart Buffalo, this panel from a warrior parade shows a horse painted with symbols denoting successful horse raids. “Warrior parades – provided the audience with another means of measuring prestige. Each detail in his drawings gives clues to the particular warrior society that the Indian belonged to. “(Tillet 1976:32)

Does the difference in orientation between the rock art panel and the other examples negate any comparison or assumption of similarity? With a symbol this simple and common my feeling is no, it is recognizable from any angle and is probably not changed by changes in orientation. Perhaps some of the other shapes and symbols in rock art had similar meanings to the artist’s who produced them. Many of Mail’s other symbols are fairly common in rock art of Colorado and the West. Keep your eyes open, it’s at least worth thinking about.


Keyser, James D.
1989    Ledger Book Art: A Key to Understanding Northern Plains Biographic Rock Art, p. 86-111, Rock Art of the Western Canyons, edited by Jane Day, Paul D. Friedman, and Marcia J. Tate, Denver Museum of Natural History and Colorado Archaeological Society, Denver.

Mails, Thomas E.
1972    Mystic Warriors of the Plains, Barnes and Noble Books, New York.

Tillet, Leslie
1976    Wind on the Buffalo Grass: The Indians’ Own Account of the Battle at the Little Big Horn River, & the Death of their life on the Plains, Thomas Y. Crowell Company, New York.

Saturday, January 4, 2014


Possible Coronado inscription. McGlone, Barker,
and Leonard, 1994, p. 74-1.

Possible Coronado inscription. McGlone, Barker,
and Leonard, 1994, p. 74-1.

In the western Oklahoma panhandle this inscription can be found on a cliff. It purports to record the passage through this area of Francisco Vasquez de Coronado y Lujan on his 1541 expedition searching for the wealthy civilization called Quivira to the east and says “Coronatto, 1541”. If authentic, it is the earliest known historic inscription in North America.

Close-up of the possible Coronado inscription.
McGlone, Barker, and Leonard, 1994, p. 74-1

“It was early fall, the time when the maize plants begin turning brown, 1540. Twenty-two summers had passed since the conqueror Hernán Cortés first stepped ashore on the mainland of Mexico, to trade he said. Now, eighteen hundred miles northwest of that dank tropical coast, a small column of helmeted Spanish soldiers marched across high, semi-arid country through arroyos, chamisa, and piñon to receive homage from the fortress-pueblo of Cicuye.
Even though they numbered not more than twenty, this medieval-looking detachment from the expedition of Gov. Francisco Vázquez de Coronado faithfully represented the conquering forces of Catholic Spain in America. The youthful captain, who wore a coat of mail and rode a horse covered with leather or quilted cotton armor, hailed his earthly Holy Caesarean Catholic Majesty in the same breath as his Heavenly Father.” (Kessell 1979:3)

Coronado's expedition by Frederic Remington.

Map of the route of Coronado's expedition. Wikipedia.

After searching the Llano Estecado of eastern New Mexico and northwestern Texas his party met a band of Native Americans that he called the Teyas, probably a Caddoan speaking group.
“The Teyas told Coronado that he was going the wrong direction. Quivira lay to the north. After more than thirty days journey, Coronado found a river larger than any he had seen before. This was the Arkansas and the spot where he reached it was probably a few miles east of present day Dodge City, Kansas. The Spaniards and their Indian allies followed the Arkansas northeast for three days and found Quivirans hunting buffalo. Coronado reached Quivira itself after a few more days of traveling. Coronado believed that there were 25 settlements in Quivira. The Quivirans were simple people. Both men and women were nearly naked. Coronado spent twenty-five days among the Quivirans trying to learn of richer kingdoms just over the horizon. He found nothing but straw-thatched villages of up to two hundred houses and fields containing corn, beans, and squash. A copper bell was the only evidence of wealth he discovered. The Quivirans were almost certainly the ancestors of the Wichita people.” ( )

Quivira, mid-1890s excavation, Archaeology, Nov.-Dec. 2013, p. 10.

Quivira - 1970 excavation, Archaeology, Nov.-Dec. 2013, p. 10.

“In the mid-1890s, the site now known as El Cuartelejo was excavated by two professors from the University of Kansas. They found the lower portion of stone walls that formed the foundation of a pueblo, inside of which were artifacts such as stone and bone tools, ornaments, and pottery sherds, some of which came from the pueblos of the Southwest. In 1970, Tom Witty of the Kansas State Historical Society excavated further, and unearthed the entire pueblo floor, hearths, and postholes. – El Cuartelejo – was a seven-room pueblo – covering about 1,600 square feet.” (Banyasz 2013:10) So, Quivira was not quite up to Coronado’s expectations. It had been built and occupied by Taos Indians in the 1600s in what is now Western Kansas, and was later occupied by a group of Picuris Indians between 1696 and 1706. (Banyasz 2013:10) So, far from the 25 cities wealthy with gold that Coronado expected, it turned out to be a small 7-room pueblo. Quite a disappointment that.

Proposed route marker from Coronado expedition. 
Bill McGlone, Ted Barker, and Phil Leonard, 1994.

Proposed route marker from Coronado expedition. 
Bill McGlone, Ted Barker, and Phil Leonard, 1994.

Some historians claim that Coronado carved the Castilian-style inscription "Coronatto, 1541" on Autograph Rock near Boise City in Cimarron County. They also ascribe a compass-like marking in the area to the same source and believe it to be a route-marker. As I stated above this would, if authentic, be the earliest known historic inscription in North America, but its authenticity is debatable. That leaves the Don Juan de Oñate inscription on Morro rock, Cibola County, New Mexico as probably the earliest known historic inscription since the Oñate inscription is agreed by historians to be genuine. But isn’t this an intriguing possibility?


Banyasz, Malin Grunberg

2013    Off The Grid, Archaeology, November/December 2013, Vol. 66, No. 6, p.10.

Kessell, John L.
1979    Kiva, Cross, and Crown, the Pecos Indians and New Mexico 1540-1840, National Park Service, U. S. Department of the Interior, Washington D. C.

McGlone, Bill, Ted Barker, and Phil Leonard
1994    Petroglyphs of Southeast Colorado and the Oklahoma Panhandle, Mithras, Inc., Kamas, UT.