Hogup Cave, Utah

Reproducing the Atlatl from Hogup Cave, Utah: A Unique Example of an Archaic Great Basin Atlatl Tradition

By Devin Pettigrew

10/8/2013   (updated 9/12/2014)

The site

Hogup Cave is located on Hogup Mountain ten miles west of the Great Salt Lake in Utah’s Great Salt Lake Desert. The cave formed by solution of the small limestone outcrop during the Pleistocene, when much of the eastern side of the Great Basin was submerged under pullivial Lake Bonneville. Evidence of ancient shorelines can be found on the mountain and in the cave. The cave consists of two large rooms that prior to excavation were linked only by a small crawl space. The opening to the cave is a large arched portal approximately 20 feet high and 30 feet across. The opening faces south allowing plenty of light into the cave.

Cultural deposits in the cave span from approximately 6,500 years BP to contact. Commonalities are observed between the Hogup Cave deposits and those of Danger Cave 72 miles to the northwest. A cultural relationship is therefore suggested for the two sites by Aikens. For the purposes of this article, it will also be helpful to note stylistic relationships between the atlatls of Hogup Cave and those found elsewhere in the Great Basin.

Excavations at Hogup Cave were carried out by C. Melvin Aikens and team between 1967 and 1968. The outer chamber, which contained an extensive looter’s trench, was excavated in the first season of field work. The two chambers were initially joined only by a crawl space, however the opening was enlarged via excavation and the second room was excavated in the second season. Excavations uncovered an extensive array of well-preserved perishable artifacts including moccasins, fragmentary components of atlatls and darts, bow and arrow components, rabbit fur robes, a bone whistle, sheep horn tools, and more. The artifacts are now located at the University of Utah’s Museum of Natural History. However, the atlatl in question along with other materials, including a rabbit net and baby moccasins, were excavated by non-professionals after the formal excavations were completed. The collectors allowed the artifact to be analyzed, and a description and photo are given by Dalley and Petersen as an addition to Aikens’ report (1970:Appendix X).

The Atlatl

Figure 1. Three views of a reproduction of the Hogup Cave atlatl, made by the author.

The Hogup atlatl is in a remarkable state of preservation and is a fairly complex conglomeration of characteristics. Photos of the original cannot be published here due to copyright; however my reproduction (Figure 1) is pretty close, and along with a comparative description of the original should suffice for general interest and research. Since crafting the reproduction from the B&W photo and information provided in Appendix X, I have seen high resolution color photos of the original. A few minor details are off, and these will be discussed below.

As you can see there is a hole in the groove that is likely accidental and happened during production. The wood at the bottom of the groove is extremely thin. This is no surprise; other atlatls from the Desert West also have holes through the bottoms of their loading grooves (e.g. Broken Roof Cave—see Garnett 2012). Burned lines around the distal end may have had symbolic value. Based on the shape of the lines themselves, it appears they were made by using a stick with a glowing ember on the end that was held against the shaft while the craftsman simultaneously blew softly on the ember (Justin Garnett, personal communication 2013).

The weight of greenish slate or shale sits in a more slanted position on the original (reasoning discussed below). I found it difficult to attach the smooth weight that I ground out of local black slate in a slanted position with wet sinew, so it came out straight on my reproduction. There are notches in the wood for the sinew at the proximal and distal bindings of the weight. The weight is broken straight across at the proximal end, and the binding has been applied over the break. There are two other sets of notches in close proximity to the proximal binding, one of which still has a remnant binding around it. An area of fresher wood below the break shows that the weight once extended nearly to the proximal most notches.

The loop is canted to correct a natural tendency of the hand to cant inward towards the head in the holding position when using a split-finger grip (Figure 2 & 3). Thus the atlatl naturally and comfortably lies flat in the holding position. Many Basketmaker type atlatl artifacts have straight loops that require a different gripping technique. These contrasts are good examples of cultural and likely individual preferences for variations on the split-finger gripping method. The loop is made of a strip of leather, folded (in half or lengthwise—I can’t tell) and wrapped around with sinew. It sits in a shallow notch carved only on the side of the loop, though the rest of the handle tapers in on both sides and is actually narrower than the notch immediately proximal to it. The attachment method for the loop on the original is a bit different than on my reproduction, with both proximal and distal ends lying flat against the shaft and tied down with sinew. Mine has the loop being secured proximally with a sinew wrapping at its center, then folded lengthwise and wrapped together. This worked but resulted in an annoying squeak at the attachment point. Cordage can be seen under the distal sinew bindings on the original and appears to be attached to the distal end of the loop. I also sewed a strong piece of sinew through the distal tab to ensure it would not slip out from under the wrappings. It is difficult to tell the exact method used to attach the loop, but it is fairly easy to make the general concept work.

The atlatl is very thin (averages 4 mm). The original is likely made of mountain mahogany (Cercocarpus ledifolius), a very hard wood. Not having access to mountain mahogany I’ve made my reproduction from black locust (Robinia pseudoacacia), which is also very hard and rigid. Thin wooden splints are bound at the loop and at the end of the handle. On the Cowbone Cave atlatl (Hester 1974b), also from the Great Basin, a bone or antler cylinder is attached to the end of the handle, most likely for pressure flaking. The Hogup Cave atlatl is missing whatever was attached in the “vice” formed by the proximal splints, which project 8mm beyond the end of the handle. I put a small bone piece in the vice. It seems to be set in very solidly. If done right I am confident the atlatl would also work well as an “Ishi stick”, a pressure flaking tool for flint knapping. The splints under the section where the finger loop is attached are more of a quandary. An additional smaller splint is visible in the color photos and appears to have been shoved under the lashings to tighten them. The larger splints were likely also added to facilitate the formation of the grip (Garnett, personal communication 2013), but placed there from the beginning. Garnett reached a greater quandary with his reproduction of the Kramer Cave atlatl (Garnett 2011). These splints are a curious characteristic of this Great Basin atlatl tradition.

Figure 2. Gripping the Hogup atlatl with the index finger through the loop in a split-finger grip is comfortable and effective.

I suggest the configuration of the atlatl as it was found represents its final functional configuration after a life of hard use; aesthetics having been set aside for functional adequacy. The atlatl weight seems to have been broken, as a purposeful adjustment or by accident, and hastily reattached, thus resulting in its current slanted position. The presence of two more sets of notches and remnant lashings may suggest previous adjustments; although the attachment of a decoration or fetish could explain the proximal most pair. Likewise the owner may not have gotten around to refitting a bone or antler bit in the vice for pressure flaking before the atlatl was deposited. The loop area also appears to have seen modification, specifically at the distal attachment point where an old binding is covered by one of fresher appearance, and cordage is haphazardly intermixed.

I have used coyote willow (Salix exigua) darts with the reproduction of the same calber as that used with the Nicolarson Cave atlatl (Figure 3) (Pettigrew 2012) and the White Dog Cave Basketmaker darts (Pettigrew 2013). The same reasoning as that used on the Nicolarson reproduction can be applied here to make a best guess about darts to use with the atlatl based on the functional characteristics of other systems that have been found with intact weighted atlatl and paired dart components—the balance point is best placed in the vicinity of the end of the handle. Despite the stylistic comparison with the atlatl fragment from Kramer Cave, a reproduction of Dart C from Garnett’s experiments in reproducing the Kramer Cave dart components (Garnett 2011) did not balance or throw well with the Hogup atlatl. It was far too light. Dart fragments from Hogup Cave are too fragmentary to support reproduction, however, they do hint at similar concepts of dart construction as those seen at Kramer Cave, with three or possibly four separate segments of varying materials that could be fitted together to make a complete dart. It is unfortunate that a whole system (dart and atlatl together) cannot be accurately reproduced, but it is comforting that the reproduction functions well with dart calibers that are common to the Desert West.

Figure 3. The author holding the Hogup atlatl at the ready. The canted loop corrects a natural tendency of the hand to cant inward in the holding position. Photograph by Ken Villars.


Atlatls from the Great Basin and peripheral regions often exhibit a similar form in the distal end, consisting of: 1) a shaft that tapers in toward the spur, 2) a spur that is set close to the distal end with little to no head-space behind it, and 3) a long, narrow, parallel sided groove, or a wide shallow groove close to the width of the atlatl shaft (artifacts include Gypsum Cave, Kramer Cave, Lovelock Cave, McClure, Par-Tee, Quiltanton Lake, Rasmussen Cave; see Pettigrew and Whittaker 2012 for references; Shearer 2013 for spatial distribution). This is in contrast with the Basketmaker form to the South, consisting of a shaft that tapers outward toward the distal end and often substantial head-space behind the spur. The latter also makes an occasional appearance in Great Basin artifacts.

The other three atlatl fragments from Hogup Cave—a possible mid-section, a proximal and a distal—suggested to Aikens that atlatls of the generalized “Mexican” type (consisting of a board body with a groove leading to a partially inset integral spur) were in use, but were not adequate to provide a full understanding of the form. With the discovery of the whole atlatl some unique variations are apparent, including the single loop at the handle. The reproduction still functions best with a split-finger grip, which is a trademark of the Mexican type. The Hogup atlatl fits best in a Great Basin atlatl tradition that is distinct from the Basketmaker type of the Southwest via the three parameters discussed above; and yet variations are apparent between the Hogup atlatl and those of the western side of the Basin approximately 330 miles away (Rasmussen Cave, Kramer Cave). The Rasmussen artifact is made for the attachment of double loops, no weights are apparent, and the atlatl is narrower and thicker. It is likely the Basketmaker and Basin atlatl types influenced each other in some cases. Further, there are variations between the Hogup atlatl and the other fragments from the same cave, but are these cultural or individual? Without a larger sample size it is difficult to tell, especially considering our lack of understanding of the temporal provenience of these artifacts.

For now, without making our heads ache further, it is best to note the relationship of characteristics based on the three parameters, suggesting a Great Basin atlatl tradition that is distinct from the Basketmaker tradition to the south, despite the likelihood of influence. It is certainly not surprising to see artifacts with variations outside of these parameters in the Great Basin as well. A better understanding of the temporal provenience of the artifacts would likely improve the situation.


Despite the lack of adequate details on the darts to support the reproduction of a fully functioning atlatl system from Hogup Cave, reproducing the atlatl can provide some useful insights into Great Basin atlatl use. Regional variations and similarities between Basin and other Desert West atlatl gear are apparent, but need more research, especially regarding temporal provenience. The Hogup atlatl functions well with dart calibers common to the Desert West. It is one of the best examples of a Basin atlatl tradition defined via three parameters in distal form, yet also shows localized variation in that form. The Hogup atlatl likely represents a very personal Archaic hunting tool deposited in a state of functional adequacy after a life of hard use, repair and alteration.


Table 1. Dimensions of the Hogup Cave Atlatl

Length overall 56.5   [centimeters]
Of spur to distal end 1.2*
Of groove to spur tip 13.5
Of hole in groove 2.3
Of handle from proximal end of loop bindings (likely includes vice projection) 8.7
Of notch in which loop sits 5.3*
Of ventral loop splint 7.3*
Of dorsal loop splint 6.5
Of vice loop splints 5
Of weight 10
Distance from distal end to top of hole in groove 9.4*
To first pair of notches (distal weight binding) 19
To middle weight binding 24 (23.5*)
To second pair of notches (proximal weight binding) 27.2
To third pair of notches (under remnant binding) 29.3*
To fourth pair of notches 30.7
Width of shaft at spur tip 2.5*
At proximal end of groove 3.5*
At middle binding of weight (not including binding thickness) 3.5*
Just proximal of last pair of notches (from distal end) 3.2*
Just above loop bindings 3*
Of loop notch 2.5*
Of handle just below loop binding 2.4*
Width of vice splints 0.6-0.7*
Of ventral loop splint 0.7*
Of dorsal loop splint 0.7
Of groove 0.8
Of hole in groove at distal end 0.4*
Of hole in groove at proximal end 0.2*
Of weight 4.2
Thickness of atlatl shaft in general 0.35-0.45cm
At loop splints 1*
At vice 1*
At spur 0.7*
Thickness of weight 0.6
Depth of groove 0.2-0.3
Diameter of loop (likely from exterior of loop) 2
Of interior of loop on my reproduction 1.6*

 Measurements are a combination of those reported in appendix X, and Adobe Photoshop measurements of Figure 1 in Appendix X resized and scaled based on the overall length, and of the scaled color photographs. My digital measurements are marked by an asterisk. When measurements consistently disagree mine are included in parenthesis.



Aikens, C. M., J. M. Adivasio, G. F. Dalley, and K. L. Petersen

1970 Hogup Cave. University of Utah Press, Salt Lake City.

Garnett, Justin

2011 Experiments in the Use of an Early Lovelock Atlatl and Dart Assemblage from Kramer Cave, Nevada (26WA196). Electronic Document, http://basketmakeratlatl.com/?page_id=75, accessed October 10, 2013.

2012 The Broken Roof Cave, Cist 1 Atlatl. Electronic Document, http://basketmakeratlatl.com/?page_id=491, accessed October 10, 2013.

Hester, Thomas R.

1974 Supplementary Notes on a Great Basin Atlatl. In Great Basin Atlatl Studies, edited by R. F. Heizer, Pp. 29-31. Ballena Press, Ramona, California.

Pettigrew, Devin B.

2012 The NV-Wa-197 Atlatl: Boatstones, Grips and Fulcrum Points on Early Archaic Atlatls. Electronic Document, http://basketmakeratlatl.com/?page_id=100, accessed October 10, 2013.

2013 Making a Basketmaker Atlatl Dart Using Modern Tools. Electronic Document, http://basketmakeratlatl.com/?page_id=819, accessed October 10, 2013.

Pettigrew, Devin B. and John C. Whittaker

2012 North American Atlatl Artifact List. Electronic Document, http://basketmakeratlatl.com/wp-content/uploads/2012/04/NA-Atlatl-Artifact-List.pdf, accessed October 10, 2013.

Shearer, Mich

2013 Distribution of Atlatls in North America. Electronic Document, https://maps.google.com/maps/ms?msa=0&msid=205393645851739092022.000488392193be41c0a2e&sll=32.876838,-118.904814&sspn=73.26438,158.027344&hl=en&ie=UTF8&ll=32.842674,-108.984375&spn=63.248992,92.965879&t=h&sourc, accessed October 10, 2013.

Comments are closed.