The Great Drought and the Middle Archaic

The Great Drought and the Middle Archaic 

by John Bradford Branney

Figure 1. Mallory Points from the Middle Plains Archaic.  

Author’s note: B.P. stands for "Before Present" and is the standard measurement convention used in radiometric dating. B.P. refers to an uncalibrated radiocarbon age with a baseline date of A.D. 1950. Measuring the amount of the radiocarbon isotope Carbon-14 in a sample from a dead plant or animal can be used to determine when it died. Since the radiocarbon isotope Carbon-14 has fluctuated through time, a correction is necessary to convert the uncalibrated radiocarbon age to a calendar date (Branney 2019). Once scientists correct the uncalibrated radiocarbon measurement, a date can be reported in calendar years, years ago, B.C., A.D., or cal BP. In my article below, I used both uncalibrated radiocarbon age and corrected calendar dates. I hope this is not confusing to the reader.    

In my last article, A High Plains Paleoindian Projectile Point Journey, I indicated that a few archaeologists believed that there was a cultural dichotomy on the High Plains sometime between 9,000 and 10,000 years ago. That dichotomy existed between the Cody Complex out on the plains and the Foothill-Mountain Tradition in the foothills and high country of the Rocky Mountains. Based on archaeological evidence, the people who made Cody Complex projectile points preferred to occupy the open plains and prairies, focusing on a bison-centric economy. In contrast, the people who made Foothill-Mountain Tradition projectile points preferred the foothills and mountains and led a more diverse lifestyle, incorporating more plants and mountain sheep into their diets. Of course, their preferences did not exclude the Cody Complex people from traipsing into the foothills and mountains on occasion, or the people of the Foothill-Mountain Tradition from visiting the plains periodically. 

Between 8,500 and 8,000 years ago, Cody Complex sites and their diagnostic projectile points disappeared from the archaeological record (Knell and Muniz 2013:13). The ecosystem was changing. Based on a postglacial climatology study in the Great Basin of North America, Antevs (1948) proposed a time period called Anathermal from around 10,000 to 7,500 B.P. (i.e., Radiocarbon Years Before Present). During the Anathermal, the melting remnants of the Laurentide ice sheet had less of an effect on the North American climate, and the polar front was migrating northward. Based on his study, Antevs suggested that during the Anathermal, the climate across the Great Basin became warmer.   

Figure Two. A chronological table based on Antevs's model of the Neothermal. The left column is Ardelean's proposal for a Transitional period (Ardelean 2016), based on the climatic modeling of E. Antevs (1955:323). 

Antevs named the time period from 7,500 to 4,000 B.P. the Altithermal. During the Altithermal, the climate became even warmer and drier than the preceding Anathermal period. The Altithermal has been called the Holocene Climatic Optimum, Hypsithermal, and Mid-Holocene Warm Period. Bonnicksen (2000:49-51) referred to the Altithermal as the Great Drought and suggested it was a period of recurring drought conditions causing regional environmental zones to shift to higher elevations. Bonnicksen stated that rainfall dropped as much as 25 percent in the Midwest and temperatures rose as much as 4 degrees Fahrenheit. Strong, dry westerly winds from the Pacific created hot and windy conditions. Extensive fossil sand dune fields along alluvial floodplains indicated that sand dunes advanced across the Great Plains. Lake levels dropped, and reliable water sources dried up. Mountain glaciers that had advanced during the last Ice Age retreated. The hotter, drier climate accelerated water evaporation and reduced soil moisture, suppressing plant growth and reducing organic carbon storage in the soil, which further compounded the situation by releasing more carbon dioxide into the air. Bonnicksen wrote that the lack of precipitation and higher temperatures affected the entire North American ecosystem and changed the landscape forever. 

What effect did the Altithermal have on human habitation on the open plains? Our 20th-century ancestors experienced it during the Dust Bowl years of the 1930s (Figure 3). The Dust Bowl years were a challenging time for land and the lifeforms. The severest drought conditions lasted only for a few years, but ravaged communities, animals, and the economy. Now, imagine Altithermal conditions over several millennia with prolonged drought conditions and sporadic, unpredictable monsoonal rains. The support structure did not exist for prehistoric people to survive such a long-term climate event, so what did they do? They adapted, or they died. Based on the lack of archaeological evidence, it appears that human populations migrated toward major river systems or to the cooler high country. Archaeological sites on the open plains during the Altithermal are rare. Sites are more commonly found at higher elevations, along mountain streams, and near the fringe areas between the mountains and the plains (Frison et al. 1976:33-34).    

Figure 3. A photograph of a farm during the Dust Bowl Years.
It was a tough time to be a farmer. 

Anthropologist William Mulloy (1958) proposed that the Altithermal climatic event caused a cultural hiatus across the entire northwestern plains. Wedel (1978:19) claimed that the Altithermal reduced the carrying capacity for prey animals on the Central Plains. Carrying capacity refers to the maximum population of a particular animal species that an ecosystem can support. An ecosystem's carrying capacity is influenced by food, water, habitat, and other resources the animal species needs to survive. Fewer prey animals on the open plains during the Altithermal resulted in a land that supported fewer humans.   

Wood (1998:38) added that the Altithermal caused deep lakes to dry up and was accompanied by extensive erosion and alluviation of drainage systems. He noted that critical prey species decreased in size during the Altithermal. Based on the discovery of fossilized sand dunes in the Great Salt Lake, Simms (2016:94) proposed that the water level reached its lowest level ever between 7,000 and 6,000 years ago during the Altithermal. The water level in the Altithermal was 13 feet below its lowest historically recorded level in 1963. He described the Great Salt Lake as a mere puddle during the Altithermal. 

Stone (2003:250) noted that in the 7,000-square-mile Denver Basin, high temperatures and low precipitation during the Altithermal led to the encroachment of cacti and woody brush and the shrinkage of short-grass prairie. Brunswig (1992:18) declared that semiarid to arid short-grass and sagebrush/yucca prairies replaced the earlier mixed tall- and short-grass ecosystems. Those changes placed plant, animal, and human communities under severe environmental stress, resulting in a substantial reduction in plant and animal biomass, particularly on the open plains. Stone added that except for a few isolated herds, bison were absent from the Denver Basin until the end of the Altithermal. She suggested that the worst drought conditions occurred between 5050 and 3550 B.C. 

From his archaeological and geological research in the Colorado mountains, James Benedict proposed two major Altithermal droughts, one from 7000 B.P to 6500 B.P., and the other from 6000 B.P to 5500 B.P. (Cassells 1997:120). Reeves (1973:1246) had a slightly different perspective on the Altithermal. He never disputed that the land was hotter, drier, and windier because it was. However, Reeves believed that the Altithermal created more shortgrass for bison to graze, and that humans adapted to the changes by shifting their subsistence strategies. 

The palaeoecological evidence for the Altithermal is compelling and paints a bleak picture for life across the Great Plains during that prehistoric time. Most researchers agree that the Altithermal brought long periods of hot, dry, and windy conditions interspersed with short, less severe conditions. Nevertheless, prolonged periods of hot weather with little precipitation dried up water sources and most likely created hardship for humans and animals. For bison and other prey animals, less forage and different water sources meant altering seasonal migrations, summer ranges, and the size of herds the land could support. That, in turn, affected human behavior. Benedict (1978:144) reviewed Altithermal Period archaeological sites in western North America, noting that nearly all of them were located in the foothills or higher country, near reliable water sources, and in zones with diverse vegetation.        

Based on archaeological and geological evidence, the climate on the plains became more hospitable between 6,000 and 5,000 years ago. Bison herds gradually returned, and humans followed. Even with a more favorable climate, life was probably rough for those first humans returning to the plains. They were not returning to a Garden of Eden. The land had been ravaged by harsh climate conditions for several millennia. It would have taken the ecosystem a long time to rebound, if it ever did completely. And the humans were returning to the open plains after generations of near absence. They would have to learn unfamiliar territory all over again through trial and error. Without knowledge of the remaining water sources, travel and trading with other tribes was risky. It is not much of a leap to assume that those Middle Archaic humans worked harder to survive than their Paleoindian predecessors. 

Figure 4. McKean Site 
projectile points.  

Signal Butte in western Nebraska and the McKean Site in Crook County, Wyoming were the first archaeological sites to define the Middle Archaic on the High Plains. Those two sites defined new projectile point types for the Middle Archaic in post-Altithermal sites. I will focus on the McKean Site, the type site for the Middle Archaic on the High Plains. The Smithsonian Institution's Missouri River Basin Survey originally recorded the McKean Site during salvage work for a future reservoir in northeastern Wyoming called Keyhole. Extensive excavations were conducted in 1951 and 1952 by archaeologist Richard P. Wheeler, and over a hundred projectile points were discovered at the site (Figure 4). Wheeler (1985:6) classified the projectile points at the McKean Site with a high level of specificity based on shape, size, and other factors. He then proposed an unprecedented nine new projectile point types from that one site. 

William Mulloy (1954) of the University of Wyoming reinvestigated the McKean Site. He interpreted the differences among the projectile points at the McKean Site as variations around a common theme. Mulloy advocated unifying the Middle Archaic cultural horizon into the McKean Complex with a single projectile point type called McKean. Wheeler (1985:9) later opined that the McKean Site should be divided into an Early Stage and a Late Stage, with each stage having widely dispersed components of distinct types and forms of material assemblages. He opposed Mulloy's proposal to combine all the components into a single archaeological complex called McKean. Frison (2014:234) reported that Wheeler and Mulloy never wavered on their differing opinions about the McKean Complex. 

Most people believe that there are three main projectile point types in the McKean Complex: McKean Lanceolate, Duncan, and Hanna. Some people believe that there is a fourth projectile point type in the McKean Complex called Mallory (Figure 1). I will not cover the Mallory projectile point type in this article, but I did cover it in a previous article (Branney 2020). There are also a few people, including me, who struggle to differentiate between Duncan and Hanna points. Taylor (2006:322) recommended combining Duncan and Hanna points under a common projectile point type called McKean Shouldered. After finding and documenting hundreds of Duncan and Hanna points, I see no value in classifying every point as either a Duncan or Hanna since many are difficult to classify. Rennie (1994:146) wrote that most McKean Complex researchers cannot even agree on the attributes that define and differentiate Duncan and Hanna points. Fortin (2015:119) found all three McKean Complex point types in her investigation of many sites from Canada to Colorado. 

Figure 5. McKean Lanceolate points surface recovered by
author in Wyoming and Colorado.  
John Bradford Branney Collection. 

Figure 5 is a photograph of five McKean Lanceolate points recovered on private land in Wyoming and Colorado. McKean Lanceolate points resurrected indented-base technology from earlier Paleoindian points such as Clovis, Goshen-Plainview, Folsom, Midland, and James Allen. However, comparing the average McKean Lanceolate point to a well-made Paleoindian point shows that they both have indented bases. That's about the only attribute that is comparable. In general, McKean Lanceolate points were randomly flaked (sometimes poorly) and were often asymmetrical in shape. From a craftsmanship and quality perspective, McKean Lanceolate projectile points could not compare with a typical Paleoindian projectile point. 

Did McKean Complex flintknappers lack the ability to craft beautiful projectile points? I do not believe that was the case. I do not believe the McKean people invested the time and effort to create works of flintknapping art as were many Paleoindian projectile points. Even though they are few and far between, I have found many beautiful McKean Complex projectile points. Perhaps McKean flintknappers stopped when they reached an acceptable level of functionality? Maybe other priorities demanded their time. We must remember that the High Plains was healing from perhaps the warmest and driest period ever in the Holocene Epoch. Herds of animals were just returning to the plains. For prehistoric people migrating onto the High Plains after the long Altithermal hiatus, surviving must have been their highest priority, and a projectile point meeting an acceptable standard was just good enough. McKean people probably received a higher return on their time investment by hunting in the field. After all, the people could not eat projectile point aesthetics. 

When I do find a well-made McKean Lanceolate point, it is sometimes difficult to differentiate it from an unexceptional Paleoindian point. At that stage, I rely on the presence of polishing along the proximal edges of the projectile point. If there is no polishing along the edges, I call it a McKean Lanceolate. In most cases, McKean Lanceolate points are quite distinguishable from their Paleoindian counterparts by the lower level of craftsmanship.  

Figure 6. Duncan, Hanna, or McKean Shouldered points
surface recovered in Wyoming and Colorado.
John Bradford Branney Collection.  

Figure 6 is a photograph of two other projectile point types in the McKean Complex, Duncan and Hanna points, or, as I previously mentioned, McKean Shouldered projectile points. Duncan points were originally described with triangular blades with slightly straight or convex edges. The stems on Duncan points were straight to slightly expanding, and the shoulders were angular to rounded. The base had a V-shaped notch. I classified points 2, 3, and 4 in Figure 6 as Duncan points. You may disagree with my classification. Hanna points were originally described with straight-to-convex blade edges, distinct shoulders, and expanding stems. The basal corners on Hanna points were rounded and lobed, and the side notches were medium depth. I classified points 1 and 5 in Figure 6 as Hanna points. Again, you may disagree with my classification, and that is exactly why I lump most Hanna and Duncan under the McKean Shouldered projectile point type. 

Figure 7. Classically made Hanna dart point found 
on June 10, 2022, on a private ranch
near Hanna, Wyoming. 

When we use projectile point types to define complexes and cultures, we find ourselves in familiar territory. Of course, the radiometric age of a component at a site is quite helpful. But for those of us walking across the ground saving prehistoric artifacts from cows, plows, and towns, radiometric ages aren't much help. Of course, there were other factors used to help identify Middle Plains Archaic sites, such as an increase in stone circles and ground stone tools, but those two factors are not as objective or definitive as projectile point type.  

Mulloy originally proposed that McKean Lanceolate, Duncan, and Hanna points were part of a single cultural system, but not everyone agreed. Metcalf (1985:181) questioned whether the same culture made all three projectile point types. Of course, it is possible that different people or cultures occupied the same sites during similar times. Based on analysis of the Signal Butte Site artifacts from Nebraska, Forbis (1985:23) observed that McKean Lanceolate flintknappers tended to use whitish chalcedony while Hanna flintknappers used more reddish jasper. He raised the possibility that the people making McKean Lanceolate points sourced their raw material from a different location than those people making Hanna points. His point was: why would the same culture source raw material from two separate quarries to make two different projectile point types? Forbis concluded that there was not one cultural complex, but two: McKean and Hanna, and that the term McKean Complex should only be applied to sites where significant numbers of McKean Lanceolate points were found. In her comprehensive study of McKean Complex sites from Canada to Colorado, Fortin (2015:119) noted that McKean Lanceolate and McKean Shouldered projectile points were found at every site that she investigated. Was that just a coincidence, or were both projectile point types made by the same people? In my opinion, it is the latter.     

William Mulloy knew that Middle Archaic people occupied the McKean Site multiple times, but he did not know that those occupations occurred over a thousand years. Kornfeld and Frison (1985:42-53) reinvestigated the McKean Site in 1983, confirming Mulloy's lower cultural unit at a radiocarbon age of 4590 ± 160 B.P., or 3390 ± 210 B.C. in calendar years. The investigators concluded that the prehistoric occupations in Mulloy's lower cultural unit spanned several thousand years, and that the McKean Site was not a single large campsite, but rather an accumulation of smaller campsites recorded over several millennia. However, the 1983 investigation did not confirm whether McKean Lanceolate and McKean Shouldered projectile point types originated from the same culture or complex. 

In her study of the origin and diffusion rate of the McKean Complex, Fortin (2015) analyzed radiocarbon dates from McKean Complex sites from Canada to Colorado. Fortin concluded that once the Altithermal ended, McKean technology diffused quickly across the plains, with the oldest McKean Complex sites in and around the Bighorn Basin of northwestern Wyoming. From its alleged origin in northwestern Wyoming, the McKean Complex spread into Montana, the Dakotas, Nebraska, Manitoba, Saskatchewan, Alberta, and eventually Colorado (Figure 8). Fortin stated that by 3220 B.P., the McKean Complex's presence was diminished, and that by 2440 B.P., the complex had completely disappeared from the archaeological record.

 Figure 8. From its origins in the Bighorn Basin of Wyoming,

the McKean Complex spread rapidly across the Great Plains.

There are various hypotheses regarding the origin of the McKean Complex. Theories have the McKean Complex originating from every conceivable direction. However, it is logical to assume that the oldest McKean Complex sites in the Bighorn Basin should be close to the origin of the complex (Fortin 2015:97). Based on early salvage work in the Bighorn Canyon of northwestern Wyoming, Husted (1969:90) noted the similarities between Pryor Stemmed projectile points from the older Foothill-Mountain Tradition and the Duncan projectile point type from the McKean Complex. Both projectile point types are remarkably similar. However, the problem with that theory is the 2,500-year-long gap between the end of Pryor Stemmed projectile points and the beginning of Duncan projectile points. What projectile point bridged that temporal gap? Figure 9 shows Pryor Stemmed points from Husted's Bighorn Canyon investigation. 

Figure 9. Husted (1969:52)

James B. Benedict (1981:87) claimed there were two mysteries that Plains archaeologists have not been able to solve for several decades. The first mystery is why the Late Paleoindian cultural complexes disappeared at the beginning of the Altithermal, and the second mystery is where the McKean Complex originated some 2,500 years later. Benedict and Olsen (1973) hypothesized that Paleoindians retreated to the high country during the Altithermal

Figure 10. Benedict and Olsen
(1973:88)

and that the McKean Complex developed from those high-altitude refuges. Benedict suggested that the Fourth of July Site, or a high-country site like it, could be the bridge between Late Paleoindian and Middle Archaic. Benedict proposed that the parallel obliquely flaked James Allen and Pryor Stemmed points found at the Fourth of July Site could have evolved technologically into the McKean Lanceolate and Duncan points found at McKean Complex sites out on the open plains (Figure 10). The radiocarbon age of the Fourth of July Site came in at around 5960 B.P. Benedict noted that the Fourth of July Site was 1,500 radiocarbon years younger than other Late Paleoindian sites and 1,000 years older than the oldest McKean Site, several hundred miles north of the Fourth of July Site. Benedict put forward a plausible theory, but the proof remains undiscovered. 

While the origin of the McKean Complex remains in speculation, several archaeologists believe that they understand its fate. Reeves (1983:80; 137-138) proposed that the Pelican Lake projectile point evolved from the Hanna projectile point, and both were part of a serial phase within the cultural tradition he called Tunaxa. According to Reeves, the Tunaxa tradition lasted for approximately two and a half millennia, with temporal phases defined by three diagnostic projectile point types: McKean, Hanna, and Pelican Lake. Reeves supported his theory by pointing out the consistency of tool types surviving across those three temporal phases. Schlesier (1994:310) agreed with that assessment, stating that many researchers believed that McKean-Hanna-Pelican Lake were serial phases of the same tradition that entered the plains from the Rocky Mountain West. If Schlesier’s assertion was correct, Pelican Lake projectile points evolved from Hanna projectile points. Foor (1982) proposed that the gradual transition from Hanna to Pelican Lake projectile points occurred around 1300 B.C., with Pelican Lake taking over completely by around 1000 B.C.

Kornfeld et al. (2010: 122-124) added that around 3000 B.P., two Late Archaic manifestations on the Northwestern Plains and Rocky Mountains began replacing projectile points from the Middle Archaic McKean Complex. The first manifestation was the widespread cultural horizon called Pelican Lake, which extended across most of the northwestern and northern plains. The second manifestation the authors named was called Yonkee, but it had a far less geographical reach than Pelican Lake. I cover that technological transition in detail in a previously published article (Branney 2025).

What is your opinion about the McKean Complex? 

   

Figure 11. McKean Complex projectile points
from Wyoming and Colorado. 
John Bradford Branney Collection. 

References Cited

Antevs, Ernest

1948  The Postpluvial or Neothermal. Bulletin of the University of Utah 38:174-184. 

1955  Geologic-Climatic Dating in the West. The Society of American Archaeology. 

Ardelean, Ciprian 

2016  The “Transitional Period”: a short terminological debate around the Pleistocene-Holocene Transition in North American prehistory. Studii de Preistorie. 2016. pp. 43-62.

Benedict, James B. 

1981  The Fourth of July Valley: Glacial Geology and Archaeology of the Timberline Ecotone.   

Research Report No. 2. Center for Mountain Archeology. Ward, Colorado. 

Benedict, James B., and Byron L. Olsen

1978  The Mount Albion Complex-a Study of Prehistoric Man and the Altithermal. Research Report No. 1. Center for Mountain Archeology. Ward, Colorado.    

Bonnicksen, Thomas M.

2000  America's Ancient Forests from the Ice Age to the Age of Discovery. John Wiley and Sons. New York. 

 

Branney, John Bradford 

2019  Radiocarbon Dating 101 – The Process. Academia.  

2020  The Middle Archaic Mallory Projectile Point in Academia. 

2025  First Light, First Blood, a Pelican Lake Story in Academia. 

Brunswig, Robert H., Jr. 

1992  Paleoindian Environments and Paleoclimates in the High Plains and Central Rocky Mountains in Southwestern Lore, Vol. 58, No. 4.    

Burroughs, William J. 

2005  Climate Change in Prehistory. The End of the Reign of Chaos. Cambridge University Press.  

Cassells, E. Steve

1997  The Archaeology of Colorado. Johnson Books. Boulder.  

Fortin, Yvonne

2015  A Spatiotemporal Analysis of the McKean Complex on the Northern Plains. Master of Arts Thesis, Department of Archaeology and Anthropology, University of Saskatchewan

Saskatoon 

Forbis, Richard G.

1985  The McKean Complex as Seen from Signal Butte, in McKean/Middle Plains Archaic: Current Research, edited by Marcel Kornfeld and Lawrence C. Todd, Occasional Papers on Wyoming Archaeology, No. 4. Laramie. 

Frison, George C. 

2014  Rancher Archaeologist. University of Utah Press. Salt Lake City.  

Frison, George C., Michael Wilson, and Diane J. Wilson

1976  Fossil Bison and Artifacts from an Early Altithermal Period Arroyo Trap in Wyoming in American Antiquity, Volume 41, No. 1.  

Husted, Wilfred M. 

1969  Bighorn Canyon Archeology. Smithsonian Institute. Washington D.C. 

Knell, Edward J., and Mark P. Muniz

2013  Paleoindian Lifeways of the Cody Complex. University of Utah Press. Salt Lake City.   

Kornfeld, Marcel, George C. Frison, and Mary Lou Larson

2010  Prehistoric Hunters-Gatherers of the High Plains and Rockies. Third Edition. Left Coast Press, Inc. Walnut Creek, CA.  

Metcalf, Michael D. 

1985  Comments on the McKean Symposium Papers, in McKean/Middle Plains Archaic: Current Research, edited by Marcel Kornfeld and Lawrence C. Todd, Occasional Papers on Wyoming Archaeology, No. 4. Laramie. 

Mulloy, William

1954  The McKean Site in Northeastern Wyoming. Southwestern Journal of Anthropology, 

10:pp. 432-460.

1958  A Preliminary Historical Outline for the Northwestern Plains. University of Wyoming Publications 22(1). Laramie.  

Reeves, Brian

1973  The Concept of an Altithermal Cultural Hiatus in Northern Plains Prehistory in American Anthropologist, New Series, Vol. 75, No. 5 (Oct. 1973), pp. 1221-1253. Wiley. 

1983 Culture Change in the Northern Plains - 1000 B.C.- A.D. 1000, Occasional Paper No. 20. Archaeological Survey of Alberta.   

Rennie, Patrick J. 

1994  McKean Complex | An Analysis of Six Middle Prehistoric Period Sites in the Northwestern Plains. Scholar Works at the University of Montana. Bozeman.   

Simms, Steven R. 

2016  Ancient Peoples of the Great Basin & Colorado Plateau. Routledge Publishing. New York.  

Schlesier, Karl H. 

1994  Commentary: A History of Ethnic Groups in the Great Plains in Plains Indians A.D. 500-1500. University of Oklahoma Press. Norman. 

   

Stone, Tammy

2003  Social Islands in the Denver Basin in Islands on the Plains, edited by Marcel Kornfeld and Alan J. Osborn. University of Utah Press. Salt Lake City.  

Taylor, Jeb

2006  Projectile Points of the High Plains. Sheridan Books. Chelsea, Michigan. 

Wedel, Waldo R. 

1978  Prehistoric Man on the Great Plains. University of Oklahoma Press. Norman. 

Wheeler, Richard 

1985  The Middle Prehistoric Period in the Central and Northern Plains in McKean/Middle Plains Archaic: Current Research, edited by Marcel Kornfeld and Lawrence C. Todd, Occasional Papers on Wyoming Archaeology, No. 4. Laramie. 

Wood, W. Raymond

1998  Archaeology on the Great Plains. University Press of Kansas. Lawrence.   

About the Author

John Bradford Branney is a Wyoming native and an author, geologist, and prehistorian living in the Colorado mountains. He has hunted prehistoric artifacts for more than fifty years and has assembled a collection of over ten thousand artifacts ranging from Paleoindian to the historic Indian tribes. Branney has written fourteen books and over one hundred articles on archaeology and geology. 

Who Dun It? Midland or Firstview, or…?

Who Dun It? Midland or Firstview, or…?

By John Bradford Branney

 

Figure 1. The 2.4-inch-long ‘point in question’ for this exercise

in projectile point typology. Red dots indicate hafting-area grinding.

“Paleoindian components are traditionally seen as distinct cultural complexes, separated in time and space, and characterized by a single type or style of projectile point.” (Sellet 2001: 60).

I have thoroughly studied the ‘point in question’ in Figure 1. That projectile point was surface recovered in Cheyenne County, Colorado. It has attributes from the Midland projectile point type from the Early Paleoindian Period, the Firstview projectile point type from the Cody Complex, and the Milnesand projectile point type from the Middle Paleoindian Period. Even though Goshen-Plainview points can be easily confused with Midland points, I eliminated them from consideration since the ‘point in question’ lacked the characteristic basal thinning strikes which were often associated with the Goshen-Plainview projectile point type.    

Figure 2. Blunted/beveled proximal edge. 

The ‘point in question’ has straight to slightly convex lateral edges with a barely discernible squarish stem created by heavy grinding of the hafting area edges. Grinding or polishing along the proximal and hafting area edges was a common practice in Paleoindian and Early Archaic times. Collectors often use it to differentiate Paleoindian and Early Archaic projectile points from later period projectile points. The proximal edge of the point is straight and blunted/beveled with tiny pressure flakes taken from both faces, giving it a wedge-like appearance. No grinding is evident on the proximal edge. In my artifact database, I labeled the raw material as a dusky red Alibates agatized dolomite, but it could be something else entirely or just a run-of-the-mill dusky red agate. Grayish bands run vertically through the raw material but are only visible under the proper lighting. The flintknapper used well-controlled transverse parallel flaking, which terminated near the middle of the biface, forming a slight medial ridge and a lenticular cross-section.

Figure 3. Cross-section from proximal to distal end
for 'point in question'.   

Firstview. On December 8, 1957, avocational archaeologist Jerry Chubbuck discovered an exposed bone bed, a Paleoindian projectile point, and an end scraper in an eroded field along a tributary of the Big Sandy Creek in Cheyenne County, Colorado.  Chubbuck brought the site to the attention of archaeologist Joe Ben Wheat from the University of Colorado Museum. At that time, Wheat was unable to investigate the site immediately because he and his crew were committed to another project. Archaeologist H. M. Wormington of the Denver Museum of Natural History suggested that Chubbuck and Sig Olsen dig a few test pits near the exposed bonebed while they waited for Wheat to investigate. Chubbuck and Olsen’s investigation yielded a treasure trove of fifty bison skulls, two dozen projectile points or point fragments, and several stone tools.

Figure 4. Projectile points from the Chubbuck and Olsen 
investigation. Note similarities to Cody Complex.   

Examples of the projectile points that Chubbuck and Olsen found are shown in Figure 4. Chubbuck (1959: 4-10) originally classified ten of the points as Eden (4), Scottsbluff (4), and Scottsbluff Variants (2). I wholeheartedly agree with Chubbuck’s original assessment after viewing the five points in Figure 4. If I found any of those points on the ground as isolated finds and before the existence of the Firstview projectile point type, I would have classified them as either Scottsbluff or Eden.     

Figure 5. Wheat (1972: 24).

Joe Ben Wheat visited the site on April 30, 1958, and Chubbuck and Olsen relinquished their digging permit to the University of Colorado Museum. After two seasons of excavation, Wheat interpreted the Olsen-Chubbuck site as a single-event bison kill that took place around ten thousand years ago. Using spears and primitive weapons, the Paleoindian hunters trapped and killed at least 190 extinct Bison occidentalis beasts in a 170-foot-long ancient arroyo.  

The hunters drove so many bison into the arroyo that the investigators found the remains of fourteen animals at the bottom of the bonebed that had never been touched or butchered at all, and another forty-nine bison that were only partially processed for meat. After subtracting 9,000 pounds of meat that was wasted, Wheat (1972: 114) estimated that the Olsen-Chubbuck bison kill produced approximately 60,000 pounds or 30 tons of usable meat. What a haul! Figure 5 is a photograph from Wheat’s investigation of two cross sections cutting perpendicular across the long axis of the ancient arroyo.  

Forty-seven artifacts were found in the bone bed itself or around the bones already weathering on the surface of the ground. Of those artifacts, twenty-seven were either complete or partial projectile points. The projectile point count appeared light for dispatching 190 or more bison. The hunters either retrieved their weapon tips from the kill site or perhaps a lot of the bison were buried and suffocated to death in the arroyo. The dominant material for the projectile points was Knife River Chalcedony (33%) from rock quarries in modern-day North Dakota.

Wheat described the points as full-bodied and lanceolate or leaf-shaped. He noted that most of the projectile points had barely discernible stems, produced by grinding or polishing the hafting-area edges. He remarked that the proximal edges of the points were predominantly straight and wedge-shaped. Wheat reported the range of the projectile points from broad points with flat lenticular cross sections to relatively narrow points with diamond-shaped cross sections. He characterized the overall workmanship as excellent. The similarity in style and the outstanding craftsmanship of the projectile points make me wonder if one or more specialists within the tribe were responsible for making most of the projectile points.

Wheat (1967: 51) originally classified the projectile points as a mixture of Scottsbluff, Eden, and Milnesand points. Five years later, Wheat (1972: 140) revised his opinion and recognized most of the projectile points as a new type he called Firstview, after a nearby town. Wheat believed that a few of the projectile points were from the San Jon point type, named after a single square-based point found associated with Scottsbluff points at the San Jon Site in New Mexico (Roberts 1942). Wheat (1972: 143) commented that if that single square-based San Jon point had been found on the central or northern plains, it would have been called an Eden point. The need for a San Jon projectile point type is openly debated in both academic and artifact-collecting circles.

 

Figure 6. Keith Glasscock, in October 1953, pointing 
where he discovered fossilized human remains.
 (Photo courtesy of TARL Archives). 

Midland.  In 1953, Keith Glasscock, an avocational archaeologist from Pampa, Texas, discovered fossilized human remains in a sand blowout six miles southwest of Midland, Texas. Glasscock also found several fluted Folsom projectile points and an unfluted Folsom-like projectile point.  Understanding the significance of his discovery, he reported his find to archaeologist Fred Wendorf. Glasscock, Wendorf, and others investigated the site and found seven more fluted Folsom projectile points and twenty-one unfluted Folsom-like projectile points. 

Wendorf and his colleagues named the site Scharbauer after the landowner and used the term "unfluted Folsom point" to describe the projectile points found with the fluted Folsom points. The investigators studied the similarities and differences between the two styles of projectile points. They compared the raw material sources for the two projectile point types. The materials were similar and easily fluted, so stone was not the conditioning factor for fluting versus non-fluting. They questioned whether the two styles of projectile point were even contemporaneous. The investigators concluded that the unfluted Folsom points were made on purpose and that they were more than just fluted Folsom point rejects. At first, Wendorf and fellow investigator Kreiger were reluctant to give the unfluted Folsom points a formal status as a type, but in a forthcoming report, they designated them Midland points, after the nearby town.

 

Figure 7. Midland projectile points from the Sharbauer 
Site (Photo courtesy of TARL Archive). 

Figure 7 is a photograph of typical Midland projectile points from the Scharbauer Site near Midland, Texas. Oilfield pipe welder and discoverer of the site, Keith Glasscock, found the Midland point on the bottom row, second from right, within three to four feet of the exposed human bones. (Wendorf et al. 1955: 44-45).  

Figure 8. The 'point in question' is hiding
out with Midland points.

Greg Perino (1985: 253) described Midland points as small, flat points resembling Folsom points but with less basal indentation, and in some cases, straight bases. Bruce Bradley (2010: 474-475) expanded on the definition by stating that Midland points had wide, but relatively flat flake scars and very flat cross sections. He stated that Midland flintknappers used finely controlled percussion flaking to shape the points, but finished with non-invasive, abrupt, marginal retouch, which narrowed the points and eliminated negative bulbs and thinning flakes. Figure 8 shows the ‘point in question’ (second from the left) in a line-up with Midland points from the author’s surface-found collection. The ‘point in question’ looks right at home, doesn’t it? Do you see similarities? Differences?    

Milnesand. Joe Ben Wheat originally suggested that Milnesand points were one of three projectile point types discovered at the Olsen-Chubbuck Bison Kill, so I felt it appropriate to mention them here. E. H. Sellards (1955) investigated a bison kill northeast of Milnesand, New Mexico, where eleven projectile points were found in a bone bed, and another twelve projectile points were found on the surface of the ground. Sellards assigned the name Milnesand to some of the points. H. W. Wormington (1957: 112) described the projectile points, “They are characterized by transverse parallel flaking, but some have a fairly pronounced median ridge. They somewhat resemble Plainview points, but the bases are square, and thinning was accomplished through the removal of more and smaller flakes in such a manner as to produce a bevel. Basal grinding is present on all specimens and often extends one-half or more of the length of the point.”

Figure 9. Drawings of Milnesand points
from Perino (1985: 255).   

Wormington noted that little was known about the age or distribution of Milnesand points, but cited possible examples in Texas, Iowa, Nebraska, Alberta, Saskatchewan, and even Alaska. Whether Wormington believed those projectile points were a distinct projectile point type called Milnesand or a variety of Cody Complex or Plainview projectile points, we will never know for sure.

Does the Milnesand projectile point have enough variation in shape and features to be considered a different projectile point type? The ‘point in question’ does have transverse parallel flaking, and it does have a square and blunted/beveled base, but those attributes also fit the Firstview and Midland type descriptions. Perino (1985: 255) suggested that Milnesand points shared similar characteristics with Plainview and Agate Basin points. Buchanan et al. (2017: 274-275) noted that Milnesand points were pretty much a single-site occurrence and that the points did not have enough variation in geometric morphometrics from Plainview points to warrant a different projectile point type. Knudson (2017: 56) suggested that the Milnesand type be retained but only as a variety of Plainview points.   

Milnesand never caught on in archaeological or artifact-collecting circles. I grew up knowing very little about the Milnesand projectile point type, and even today, Milnesand is barely mentioned, if at all, in books about High Plains archaeology. Based on what I have read and seen, Milnesand is one of those projectile point types that might better fit into a more widely accepted category, such as Plainview.

Final Reflections

As you can see from this and other articles, projectile point typology is not a clear-cut science. In fact, it can be more art than science. In most cases, the attributes of a Paleoindian projectile point are so well represented that identifying the type is a ‘no-brainer’; in other cases, the point has common features from more than one type. That makes classifying difficult, if not impossible. The ‘point in question’ has common attributes from at least two projectile point types. 

Wheat (1972: 140-152) did not decide haphazardly to create a new projectile point type called Firstview. He realized that Paleoindians were not working from blueprints and that the templates they were using to make projectile points were much less rigorous than our own. He recognized that before creating a new projectile type, he needed to objectively review the similarities and differences between different projectile point types using raw material sources, flintknapper skill levels, geographic distributions, and relative ages. Wheat carefully analyzed Scottsbluff, Eden, San Jon, and Milnesand points from different sites and compared them with the Olsen-Chubbuck points. Eventually, he concluded that a new projectile point type was in order. His analytical approach to ensuring the correct decision cannot be questioned.  

There was only a single radiocarbon date available at the Olsen-Chubbuck Site, and it was obtained from bone collagen: 10,150 ± 500 B.P. (Holliday et al. 1998: 448). However, the investigators wrote that they were uncomfortable with that uncalibrated radiocarbon age because bone was difficult to date, and it was a lone sample with a very large standard deviation. The same investigators reported Firstview at other southern plains sites and established an uncalibrated radiocarbon age between 9400-8300 B.P, still within the range of younger Cody Complex sites but very young for both Milnesand and Midland.    

I do not believe that a Firstview projectile point type was necessary. In my opinion, the projectile point attributes from the Olsen-Chubbuck Site were well within the morphological and temporal range of Scottsbluff and Eden types. Of course, the Olsen-Chubbuck points were slightly different than other Scottsbluff and Eden points found elsewhere, but morphologically they were in the same ballpark. It is my contention that there is a high probability that one or two individuals involved in the bison kill made most of the Olsen-Chubbuck projectile points. That would explain the consistency in form, the impressive flintknapping ability, and the small features that differentiated Olsen-Chubbuck points from Cody Complex points found at other sites. For example, I would be hard pressed to find any significant differences between the Olsen-Chubbuck projectile points and the Scottsbluff and Eden points found at the Horner Site, the type site for the Cody Cultural Complex (Bradley and Frison 1987: 199-231).

After waffling back and forth for several years on whether the ‘point in question’ was Cody Complex or Midland, I settled on documenting it in my artifact database as a Firstview point from the Cody Complex in the Middle Paleoindian Period. I did that even though I made the case for not needing a Firstview projectile point type in the last paragraph. Hypocritical, I know, but since we have a Firstview projectile point type, it might as well be used. I made my decision based on the barely discernible Cody-like stem that the flintknapper created by grinding in the hafting area. Other evidence included its squarish base and the transverse parallel flaking. Could it be a Midland point? Absolutely. A Milnesand? Perhaps. We will never know for sure, but quite frankly, does it really matter? Paleoindians living very similar lifestyles made what we call Firstview, Midland, and Milnesand points, and they would probably never believe that ten thousand years later, we would be scrutinizing and cubby holing their tools and weapons into different categories.      

 

Comparisons of artifact types can, and will, demonstrate morphological differences, not necessarily cultural ones.

(Sellet 2001: 61)

 

References Cited

Bradley, Bruce A.

2010   Paleoindian Flaked Stone Technology on the Plains and in the Rockies. In Prehistoric Hunter-Gatherers of the High Plains and Rockies. Left Coast Press. Walnut Creek.

Bradley, Bruce A., and George C. Frison

1987  Projectile Points and Specialized Bifaces from the Horner Site. In The Horner Site – The Type Site of the Cody Cultural Complex. Academic Press. New York.    

Buchanan, Briggs, Michael J. O’Brien, and Mark Collard

2017  A Geometric Morphometrics-Based Assessment of Point Types on the Southern Plains during Plainview Times. In Plainview – The Enigmatic Paleoindian Artifact Style of the Great Plains. University of Utah Press. Salt Lake.    

Chubbuck, Jerry

1959  The Discovery and exploration of the Olsen-Chubbuck Site (CH-3). In Southwestern Lore 25.  

Holliday, Vance A., Eileen Johnson, and Thomas W. Stafford Jr.

1998    AMS Radiocarbon Dating of the Type Plainview and Firstview (Paleoindian) Assemblages: The Agony and the Ecstasy. In American Antiquity, 64(3), 1999, pp. 44 41454.

Knudson, Ruthann

2017  The Plainview Assemblage in Context. In Plainview – The Enigmatic Paleoindian Artifact Style of the Great Plains. University of Utah Press. Salt Lake.    

Perino, Gregory

1985  Selected Preforms, Points, and Knives of the North American Indians -Volume 1.

Roberts, Frank H. H., Jr.

1942  Archaeological and Geological Investigations in the San Jon District, Eastern New Mexico. In Smithsonian Miscellaneous Collections, Vol 103, No. 4.

Sellards, E. H.

1955  Fossil Bone and Associated Artifacts from Milnesand, New Mexico. In American Antiquity, Vol. XX, No. 4, pp. 336-344.    

Sellet, Frédéric

2001  A Changing Perspective on Paleoindian Chronology and Typology: A View from the Northwestern Plains. In Arctic Anthropology, Vol. 38, No. 2, pp. 48-63.

Wheat, Joe Ben

1967  A Paleoindian Bison Kill. Scientific American 216: 44-52.

1972  The Olsen-Chubbuck Site – A Paleo-Indian Bison Kill.  In American Antiquity, Volume 37, Number 1, Part 2.  

Wendorf, Fred, Alex D. Krieger, and Claude C. Albritton

1955  The Midland Discovery. Greenwood Press. Westport.

Wormington, H. M.

1957  Ancient Man in North America. Denver Museum of Natural History, Popular Series No. 4.      

About the Author

 

John Bradford Branney is the author of the historical fiction series, the SHADOWS on the TRAIL Hexalogy, adventures that took place on the High Plains around 12,600 years ago. Life was difficult for my Paleoindian characters. Every day was a toss-up between surviving and dying. The characters face all the social and physical struggles of being human, plus unpredictable acts of nature, mega predators high up the food chain, adversarial enemies, and the constant search for the next meal. Join me and my characters on an odyssey you will never forget.