Do projectile points get cold? An experimental approach examining composite and stone projectile technology

Abstract

Researchers have noted that cold temperatures may have had an impact on hunter-gatherer decisions concerning raw material selection for projectile points. This line of reasoning has been used to explain the phenomenon of projectile points of different materials occurring during the same time period in archaeological contexts that exhibit extreme seasonality. Cold temperatures are assumed to affect cryptocrystalline brittle solids adversely, whereas organic and composite projectiles are more resilient. Here, an experiment was designed to test the brittleness of stone and antler composite projectile points subjected to different temperatures. It was demonstrated that cold temperatures do not impact projectile point brittleness. However, differences in projectile raw material type were found to fracture differentially suggesting other probable factors associated with projectile point raw material selection.

Keywords

Alaska experimental archaeology projectile points microblades temperature log-linear analysis

Introduction

Basic economic thinking weighs the costs and benefits of certain aspects of a system to facilitate a solution that has the most desirable outcome. Archaeologists frequently use this line of inquiry when comparing different technologies used in the past for similar tasks in an attempt to understand the preference of one technology over another (e.g., Bettinger et al., 2006; Bleed, 1986; Fitzhugh, 2001; Hayden et al., 1996; Schiffer, 2011; Schiffer and Skibo, 1987). In 2002, Elston and Brantingham utilized an economic approach to compare the performance characteristics of organic, composite, and stone projectile points from the perspective of tool design and risk management (Table 1). Of particular interest in their study is the suggestion that stone projectile points are more brittle in cold temperatures than organic and composite organic projectile points. Here we employ a controlled experiment with replication to test the affect of cold temperatures on composite and stone projectile technology and the durability of the point types.

Table 1.

Characteristics of organic, composite, and stone projectile points based on ethnographic and experimental data (after Elston and Brantingham, 2002).

Variables	Organic points (Osseous)	Composite points	Stone points
Morphology	Limited morphology	Wide range	Variable morphology
Time investment	Moderate manufacture investment	High manufacture investment	Low manufacture investment
Tissue damage	Causes less tissue damage	Causes great tissue damage	Causes great tissue damage, bleeding
Durability	Highly durable: unaffected by cold rarely damaged rarely broken	Moderately durable: flexible technology, if the composite microblades fail, the antler can be relied on as a backup	Less durable: brittle in cold easily damaged break on miss shatter with contact
Maintenance	Easily repaired/resharpened	Easily repaired or difficult in some cases	Difficult or impossible to repair
Use life	Reusable	Composite microblades can be replaced	Expendable
Attachment	Less likely to be used with foreshaft	Less likely to be used with foreshaft	More often used with foreshaft

Cold temperature, being a catalyst for stone projectile breakage, has been passively mentioned in various temporal and spatial contexts including late Pleistocene/early Holocene Eastern Beringia (Adams, 2018; Hirasawa and Holmes, 2016; Rasic, 2008; Robinson, 2008; Wygal, 2011) and the Upper Pleistocene of Western Europe (Pétillon et al., 2016). Both of these periods and areas exhibited severe winters with very cold temperatures. Organic or composite projectile points may have been selected over stone projectile points due to the assumed elevated brittleness of stone in cold temperatures. However, even though organic and composite projectile points may be more resilient to colder temperatures, the manufacture time associated with them is substantially greater compared to stone (Guthrie, 1983a, 1983b; Knecht, 1997a; Waber, 2011). It is not the purpose of this study to isolate other probable variables that may have structured projectile point selection, such as raw material availability (Robinson, 2008; Wygal, 2009), prey type (Buchanan et al., 2011; Cotter, 1938; Graf and Bigelow, 2011; Haynes, 1964; Potter, 2011; Wood and Fitzhugh, 2018), propulsion mechanism (Ackerman, 2011; Ames et al., 2010; Clarkson, 2016; Dixon, 2013; Hughes, 1998; Potter, 2011; Shott, 1997; Thomas, 1978), climate and ecological factors (Wygal, 2011), and land use patterns (Potter, 2011; Wygal, 2018). Instead, we focused on one variable (extreme cold) and evaluated its contribution to projectile point fracture across multiple raw material types.

Sparse ethnographic data have been used to support the assumption that cold temperatures impact stone projectile points adversely. Of Ellis’s (1997) extensive review of the ethnographic literature associated with projectile points, there seems to be only one instance in which the brittleness of stone projectile points being affected by cold weather is mentioned. Upon conducting ethnographic research with the Nunamiut of northern Alaska, Binford (1979: 262–263) records:

The old men explained that the difference between “deer” and “bear” arrows was not so much that each was exclusively used for the two types of game but that the antler points were used more commonly during months of freezing weather, and stone when it was warm. Stone projectiles were very easy to break and were unreliable to carry. Because they “cracked sometimes from just being rubbed together in the quiver” when it was very cold.

Here, three hypotheses were experimentally analyzed to provide insight into the relationship between cold weather and projectile point type as well as overall durability:

H1: Antler composite technology is more durable in cold temperatures than stone bifacial technology.

H2: Antler composite technology is more durable than stone bifacial technology independent of temperature.

H3: Lanceolate stone biface durability varies by raw material type.

Material and methods

This experiment was conducted using 120 lanceolate stone projectiles made of obsidian (n = 40), chert (n = 40), and rhyolite (n = 40) and 20 composite projectile points made from antler (Figure 1). The obsidian used in this study came from the Glass Buttes source in Oregon. The rhyolite came from central Oregon near Madras, and the cherts came from Texas, Indiana, and Oregon. Manufacture times differed for the points depending on the shape and thickness of the blank but on average an obsidian point took 50 minutes, a chert point 65 minutes, and a rhyolite point 70 minutes. The assemblage was created by experienced knappers and the points had a mean length of 78 mm, mean width of 24 mm, and mean thickness of 8.5 mm. This range of point size was determined based on metric attributes from the Mesa Site in Alaska (Bever, 2000; Kunz et al., 2003). None of the raw materials in this study were subjected to heat treatment.

Figure 1.

Example of composite antler and stone projectile points used for the study.

Composite projectile points were manufactured to be ∼70 mm long with a width of ∼10 mm and tapered with a thickness of ∼10 mm at the base to ∼5 mm at the tip. The composite grooves were at a depth where the outer layer of the antler contacts the spongy core. Measurements of the points as well as the depth of the grooves were in part based on data from the Trail Creek Caves site in Alaska (Ackerman, 2011; Lee and Goebel, 2016). Twenty composite projectiles were manufactured using dry antler tines from whitetail deer and microblades of chert and obsidian. The process of manufacturing the organic element of these projectiles proved to be quite time consuming as the antler segments had to be straightened, shaped, and grooved. Antler is a hard material to work and it took experimentation to figure out which tool worked best for manufacture. An angle grinder, Dremel tool, file, and belt grinder were all tried but were not efficient in manipulating the materials. After trial and error, the most effective tool to thin the antler blank was a horse rasp, which has a bar with sharp teeth and cuts more coarsely than a file. Horse rasps are traditionally used to remove the excess wall from a horse’s hoof and represent an aggressive abrading tool. The initial blanks were longer than the specifications, and a Dremel tool with a cutting bit was used to cut the antler into similar lengths. The natural curve makes it hard to straighten which was only facilitated after soaking the antler for a 24-hour period in water. The antler was thinned in a vice using the horse rasp and then grooved and sharpened using the cutting bit and grinder on a Dremel tool. Not taking the soaking into account the straightening, thinning, and grooving of the composite projectile points took approximately 75 minutes per point. This was done with the aid of modern power tools and the process would probably have taken much longer to complete without the use of power equipment. Even with the use of power tools manufacturing symmetrical tips and grooves was difficult.

The manufacture of the microblades was facilitated through soft hammer percussion on wedge-shaped cores and the lengths were based on microblades from the Dry Creek Site in Alaska (Powers et al., 1983). Microblades were manufactured from obsidian and chert with a vice used to snap the proximal and distal end off the microblade to facilitate a uniform flat surface that would fit into the grooved antler implements. After adequately shaping the microblades to fit into the grooves, two were adhered into place using Elmer’s glue which is protein based. Instead of placing microblades along the entire length of the groove, only a pair was used due to the notion that having a full row would not make a difference in the results of the temperature study.

Both the stone and composite projectiles were attached to a 4-cm wooden dowel, 6.35 mm in circumference, used as a foreshaft with a tip that was sharpened using a standard pencil sharpener to facilitate a connection to the projectile (Figures 2 and 3). Loctite® general purpose repair putty was used to haft the foreshaft to the projectiles. The putty was heated to mix the epoxy resin and hardener to facilitate a strong bond that is temperature proven in the long term from –27°C to 149°C. It has a quick dry time, is easily moldable, and will not crack. After connecting the projectile points to a foreshaft using the putty, the sharpened end was inserted into a carbon arrow shaft with plastic fletching, commonly used in modern bow hunting applications. This connection allowed for one arrow to be used multiple times in concert with the hafted projectile.

Figure 2.

Experimental lanceolate projectile points hafted to wooden dowel foreshaft.

Figure 3.

Experimental composite projectile point hafted to wooden dowel foreshaft.

A calibrated crossbow propulsion apparatus with a draw weight of 22.7 kg was employed to launch the projectiles (see Clarkson, 2016) (Figure 4). The propulsion device used was a crossbow mounted to a stationary table using two vices to ensure the same position for every shot with a third vice holding a caliper release grip with a trigger mechanism ensuring minimal human error in the shots and perfect accuracy. A laboratory in the basement of College Hall at Washington State University was used to control for variables such as wind and temperature that may influence the flight of the projectile.

Figure 4.

The propulsion device in the firing position and the target.

The calibrated crossbow was used for this experimental design because it allowed for complete control over the variable of propulsion. It is not intended to be an analogy for propulsion mechanisms used by hunter-gatherers of the past but as a tool to examine the breakage of projectile points. Admittedly, the draw weight of the calibrated crossbow used in the experiments is similar to that of a bow and arrow (Cattelain, 1997). This may be a shortcoming of this study but opens the door for other research utilizing different propulsion mechanisms to examine the breakage of points.

Targets were individually constructed for each shot and consisted of a small plastic bag with a piece of canvas, three pieces of cardboard, and a piece of 1/8 inch plywood inside the bag with a thick piece of foam behind the plastic bag. The matrix of the target simulated a variety of different materials that were used as proxies for different types of hardness that may be encountered in a real world situation. These materials range in property from soft to hard such as an animal’s hide, muscle, and bone. This is not a direct analogy to the natural world but provided a simulated, controlled, way of testing projectile point breakage through various material types. The target was situated in a wooden box that was constructed to ensure the collection of pieces that break on impact. To obtain accurate and consistent shots, the target was positioned relatively close at 4 m from the tip of the projectile to the actual target.

Experimental overview

Temperature study

The population of stone projectile points was divided into two groups of 20 for each material type, and composite projectiles into two groups of 10. The stone and composite projectiles were then subjected to two temperatures: –30°C (extreme freezing) and 15.5°C (the controlled room temperature of the laboratory). To simulate freezing conditions, each completed projectile was placed in an industrial freezer for a one-week period to bring the external and internal temperature down. After this cycle, the projectile was transferred to a cooling bath containing dry ice and 99% isopropyl alcohol. It is of note that the cooling bath may have served as a desiccant to the materials and completely removed the moisture content; however, this would be difficult to fully test. The cooling bath has a minimum temperature of –77°C, well below the target temperature ranges for this experiment (Figure 5). This simulation served to test a gradual temperature shift as opposed to a rapid shift that may occur when projectiles are stored in a warm place then exposed to a drastic temperature change.

Figure 5.

The propulsion device in the firing position and the target.

The projectile stayed in the cooling bath for approximately 5 minutes to ensure thorough cooling. This methodology is the most controlled, standardized way of cooling the projectiles and allows for an extended deep freeze and then a flash cooling. If the experiments were conducted outside, there would be too much variability in temperature for adequate control. Next, the projectile’s foreshaft, a wooden dowel, was inserted into a carbon arrow shaft while the projectile was still in the cooling bath. The projectile was then taken out of the bath, quickly mounted on the propulsion device, and a noncontact instant output infrared thermometer with a laser pointer measured the projectile’s temperature. Once the desired temperature was achieved, the projectile was fired only once making contact with the target before warming up.

Durability study

The second portion of this study was concerned with the overall durability of the projectile points without using temperature as a variable. The durability study assumed that a point that can withstand multiple impacts were advantageous over those that break after a few target impacts. Projectile points throughout the process of impacting a target would acquire three types of damage: macroscopic, microscopic, and internal. Macroscopic damage was visible without the use of a hand lens, and an example of this would be a point snapping in half. Microscopic damage would include small flake detachments or surface cracks that might only be detected with a microscopic apparatus. Lastly, internal damage was something that occurs inside of the projectile with the stresses of being propelled into a target (see Pargeter, 2011). If a projectile had inclusions inside, it was more likely to acquire internal damage as opposed to a point that had a more homogeneous makeup. For this study, macroscopic damage was the only type examined, but that is not to say that both microscopic and internal damage would not have had an impact on how projectile points break.

To facilitate this experiment, six projectile points from each population of chert (n = 6), rhyolite (n = 6), obsidian (n = 6), and composite (n = 6) that survived the temperature study were subjected to multiple shots until breakage occurred or 10 shots had been reached. The 10-shot cutoff was arbitrarily determined and these projectiles were propelled at room temperature. No maintenance was performed on the projectile points and they were utilized until they could no longer function as projectiles.

Experimental results

Temperature study

Table 2 provides the breakage data for the experiments conducted with each projectile point subjected to a single shot. Stone lanceolate projectile breakage was variable and differed between raw material types. Of the obsidian projectiles that were exposed to cold temperatures, 10 exhibited breakage and 10 did not (Figure 6). In addition, 8 warm temperature obsidian points broke, while 12 did not. The entire population of chert points survived the cold temperature experiment with no breakage, while 2 broke and 18 did not when subjected to warmer temperatures. Also, 3 rhyolite projectiles broke during the cold weather test while 17 did not, and 6 broke during the warm temperature test while 14 did not. Composite point impacts were more complicated. The antler component of the composite points always survived impact. However, the inset microblade component consistently fractured when contacting the wooden layer of the target, leaving a lateral segment of the microblade embedded in the antler groove (Figure 7). Even though the microblade component failed, the antler portion of the projectile still proved to be a lethal projectile that did not fail.

Table 2.

Temperature breakage data for projectile points.

Point type	Cold temperature, count (%)		Warm temperature, count (%)
Point type	No point break	Point break	No point break	Point break
Compositea	10 (100)	0 (0)	10 (100)	0 (0)
Rhyolite	17 (85)	3 (15)	14 (70)	6 (30)
Chert	20 (100)	0 (0)	18 (90)	2 (10)
Obsidian	10 (50)	10 (50)	12 (60)	8 (40)
Total	57 (81)	13 (19)	54 (77)	16 (23)

^aMicroblades did break but total failure did not occur.

Figure 6.

Obsidian projectile propelled after cooling.

Figure 7.

The osseous implement of the projectiles always survived impact but the composite microblades sheared off while coming into contact with the wooden target, often times leaving a serrated-like edge.

Since the results of breakage were somewhat variable among different raw material types and warm and cold temperatures, a Fisher’s exact test was used (p = .05). This type of statistical analysis was more accurate for small sample sizes than other tests of independence, which was ideal for this instance because cell count does not impact the results. In the case of obsidian (p = .7512), chert (p = .4872), rhyolite (p = .4506), and composite (p = 1.0000) projectile points, there was no statistically significant association between temperature and breakage. These data support the null hypothesis and indicate that for each different projectile point type in this experiment, temperature did not affect breakage.

To further explore these data, a log-linear analysis was performed, which was useful for examining a statistically significant relationship between multiple variables. A log-linear analysis can be complicated in examining relationships due to the multiple outcomes generated but was the best statistical device for examining the data presented in this study because it actively examined the relationship of three variables. This analysis looked at the different associations between breakage, temperature, and point type and tested the statistical significance while providing various models. Table 3 presents the interaction variables, the different models, and the results of the analysis.

Table 3.

Results of log linear analysis.

Model	Interaction variables	G ²	df	p
A	Breakage and temperature and point type	33.84	10	.0002
B	Breakage and temperature	0.38	1	.5376
C	Breakage and point type	29.26	3	<.0001
D	Temperature and point type	0	3	1
E	Breakage and temperature with point type held constant	4.58	4	.3332
F	Breakage and point type with temperature held constant	33.46	6	<.0001
G	Temperature and point type with breakage held constant	4.2	6	.6496

The results of the log-linear analysis, like the Fishers’s exact test above, statistically demonstrate that there was no association of temperature with breakage or projectile point type. Model A, a three-way interaction, showed a significant association between the three variables. As such, the model was parsed out to further examine two-way interactions. Model B again confirmed no association between breakage and temperature. Model C showed a highly significant association between breakage and point type. Model D was not valid because there are zeros associated with the input sample, resulting in a zero goodness of fit value. Models E, F, and G look at two-way interactions with the third variable being factored out. In this case, Model E showed no significant association when breakage and temperature were compared and point type was held constant. Model F, like Model C, showed a highly significant association between breakage and point type. Lastly, Model G showed a nonsignificant association between temperature and point type when breakage was held constant. In conclusion, temperature had no impact on projectile point type or breakage. However, this is not to say that temperature has never been the catalyst for a point breaking, but results of this experiment suggests that temperature was not a variable that affects projectile points breakage.

Durability study

A total of 24 projectile points, 6 each of obsidian, chert, rhyolite, and antler were propelled multiple times to test durability before breakage. Of the stone projectile points, rhyolite proved to be the most resilient to breakage with a range of four to seven shots and a mean of 5.7 before breakage. This was followed by chert with a range of three to seven shots and a mean of 5.2. Obsidian proved to be the least resistant to breakage with a range of two to six and a mean of 4.2 (see Table 4). The differences in chert and rhyolite are particularly interesting given that during the temperature study only two chert projectile points broke, while nine of the rhyolite points broke. Only two of the antler points failed after 9 strikes, whereas the rest were retired after 10 strikes without breakage (Figure 8). It is of note that the composite projectile points exhibited the loss of microblades during the previous study; however, this does not constitute total point failure as the antler implements still penetrated the target with ease.

Table 4.

Number of target strikes before point failed.

Obsidian	Chert	Rhyolite	Composite point
6	7	6	10
5	4	7	9
6	7	5	10
3	4	5	10
2	3	4	10
3	6	7	9
Mean = 4.2	Mean = 5.2	Mean = 5.7	Mean = 9.7
SD = 1.57	SD = 1.57	SD = 1.11	SD = 0.471

Figure 8.

Composite projectile after nine shots.

An analysis of variance (ANOVA) test, used for testing the differences of means, was used to examine the projectile point breakage data further. The results indicated that there was a significant relationship between the variables (F = 18.26; df = 3; p ≤ .0001) (Table 4).

A Tukey Honestly Significant Difference test was undertaken to more specifically check the pairwise comparison among the means. The ANOVA above showed that there was a significant relationship between the number of strikes and raw material but did not show if there was a significant difference between specific material types and how they break in an experimental context. The Tukey test demonstrated that there was no significant relationship between obsidian and chert (nonsignificant), obsidian and rhyolite (nonsignificant), and chert and rhyolite (nonsignificant). When comparing stone projectile points there was a significant difference between chert/composite (p < .01), obsidian/composite (p < .01), and rhyolite/composite (p < .01). These results indicated that there was no significant difference in durability when comparing chert, rhyolite, and obsidian. However, there was a significant difference when comparing chert, rhyolite, and obsidian to composite projectile points. These results showed that antler composite points were significantly more durable than stone projectile points.

Discussion

The data above suggest that there was no significant relationship between temperature and point breakage. A unique and an unanticipated result was that the microblades mostly sheared off after the first shot with the composite projectiles. This may be due to utilizing a composite target with plywood as opposed to an actual animal carcass for the experimental design. It has been demonstrated in other experiments that microblades often fail while penetrating hide (Pétillon et al., 2011) and bone (Waber, 2011). However, this is not considered a total failure as the projectile could still easily penetrate the target and be of use even without the microblades in a hunting situation. It is feasible to think that a hunter would carry a pouch of microblades, or a small core that could be used to easily replace and maintain composites that were broken off or knocked out to re-arm the antler projectile. Even though the microblades failed, the utility of the point survives because of the durable antler element allowing for a very capable killing implement to still exist. No maintenance was performed in this experiment on either composite points or lanceolate points. However, it is conceivable that ancient hunters may have maintained projectile tips by re-arming microblades or retouching the stone projectiles to utilize these tools for a longer period of time. That being said, this was not the purpose of this particular experiment. We explored how long each projectile would hold up until total failure occurred, as would happen with a raw material shortage or a hunting event when a point is being used multiple times. Even though the microblades failed, the composite projectile was more durable than rhyolite, obsidian, and chert projectile points. So, what is the point of using inset microblades if they readily break off upon impact? Composite projectile points represent a redundant technology in which if the microblade component fails the user still has a reliable tool for dispatching animals. Clarkson and Shipton (2015) experimentally showed that when using a composite tool, such as a sickle, the tool was still useful even after losing composites. Myers (1989: 87) proposed that, “it is not possible to anticipate…the timing, location and relative abundance of resources and, secondly, it is difficult to anticipate levels and periods of technological demand” thus needing a weapon system to function highly and efficiently at short notice even if an element fails. Ideally, there would be time in a situation to re-inset microblade elements; however, in the case of mass hunting (e.g., multiple animals in a snow bank or the water), a redundant tool would be of great use (Churchill, 1993). Clarkson et al. (2018:184) note that microblades can be reattached quickly by heating the residual adhesive left inside the groove of an osseous implement. However, in a chaotic situation such as mass hunting a few minutes does not exist, as multiple animals would have been available for procurement. Firstly, the composites provide a lethal cutting edge that would cut through hide, vessels, and arteries and upon the failure of that element and the best chances of a first kill the bone/antler element of the point would plausibly be useful for dispatching more animals. The redundancy of composites can also be looked at in regard to raw material use. If it is a season of minimal tool stone availability (e.g., winter), a composite point and an excess of microblades would provide for an ideal tool and still be useful if there was not access to stone in raw form and the surplus of microblades was depleted (e.g., Wygal, 2009, 2010). This study supports that temperature does not impact projectile points; however, it is important to note the minimal sample size. This should serve as a platform for future studies and does not imply that temperature was not the catalyst for the breakage of stone points in some instances.

When comparing rhyolite, chert, and obsidian, it would seem likely that there would be a significant difference in the number of shots each type of raw material could survive. For instance, rhyolite is a harder raw material to manufacture a point from than obsidian. This thinking leads to the notion that there is a correlation between how hard a stone projectile point is to manufacture and durability (Bleed, 1986; Iovita and Sano, 2016; Knecht, 1997b; Torrence, 1989). Qualitatively it was demonstrated that rhyolite, followed by chert, then obsidian can survive the most number of shots. However, quantitatively there was not a significant difference between the rate of breakage for these three raw material types.

Our experiment looked primarily at two factors associated with projectile point fracturing, extreme cold temperature and impact durability. There were a whole host of other factors that could have been studied and we encourage others to embrace some of those factors. For instance, we did not experiment with propulsion mechanisms or propulsion velocity. Propulsion velocity may be critical for prey type or hunting distance. In this study, our microblade inserts were firmly glued and as such, did not slide out of composites after impact (they broke off). But what if composite microblades were designed to slide out after impact to promote internal hemorrhaging of prey animals? Is it efficient to re-arm composite points during a hunting excursion? We have already shown that antler is more durable than stone projectile points, but is it more efficient to use the same antler point or points that are re-armed multiple times as opposed to making many relatively time efficient stone projectile points? Certainly, this can be assessed with an experimental study as well.

Conclusions

In this study, we utilized an experimental approach to investigate the impacts of cold temperature and durability for different raw material types made into projectile points. It was found that cold temperature was not a significant factor in the breakage rate of projectile points. Our study also showed that composite antler projectile points, even with the loss of microblade composites, were more durable than any type of stone projectile point. This shows a potential trade-off between tool production investment and longevity. Clearly, there are multiple variables that might have acted on the type of projectile point used and it is not feasible to select or identify one variable as more important than others. Our experiments found that extreme cold and warm temperatures had no impact on raw material selection when considering point breakage. This is worth additional discussion. The idea that composite projectile points are more durable than stone points in cold temperatures was suggested first by Elston and Brantingham (2002) with no experimental data supporting the claim. They posed that “stone points apparently become more brittle and are easily damaged in severe cold” and that “organic points were…used in cold weather (when stone points are brittle and easily broken)” (Elston and Brantingham, 2002). This assumption has been referenced in many contexts in which there are multiple projectile point raw material types associated with cold temperatures such as late Pleistocene/early Holocene eastern Beringia (Wygal, 2011) and the Upper Pleistocene of Western Europe (Pétillon et al., 2016). Here we suggest that temperature did not have an impact on decision-making concerning projectile point production choice. However, our experimental results do reveal that composite antler points are more durable than stone projectile points and may have been a factor in the choice of projectile raw material types. We also note that other factors such as hunting strategy, prey type, and propulsion mechanism may have had a role in projectile point production as well (see Adams, 2018).

Experimental studies often utilize unique approaches to undertake research questions. These approaches may or may not be agreed upon by other researchers as representing the “best approach” but should provoke other thinking on the matter for future experiments. Science is a result of trial and error and only with this can advancement in the field occur. We hope that these experiments provide a platform for more cold temperature exploration regarding raw material use.

Footnotes

Acknowledgments

Many thanks to Colin Grier, Josh Reuther, Shannon Tushingham, Sam Coffman, Doug McDonald, Craig Lee, Jeff Rasic, Dale Sexton, Ken May, Bryan Wygal, Kate Krasinski, Chuck Holmes, and the late John Cook for taking the time to discuss this project. Dan Stueber and Steve Allely provided the projectile points for this study, and they were so stunningly done it was sad to break them. Lastly, a thanks to the phenomenal editing of Anthony Boldurian, Editor of North American Archaeologist, and the peer reviews by John Blong, and one anonymous reviewer. Any errors are solely our responsibility.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The completion of this research was facilitated through the aid of the Department of Anthropology at Washington State University, a Don Crabtree Scholarship, and the National Science Foundation through a Doctoral Dissertation Improvement Grant (#1544558).

Author Biographies

Jacob S Adams earned his PhD in Anthropology from Washington State University. He is currently a forest service archaeologist in Delta, Colorado. His research interests include lithic technology, experimental archaeology, and hunter-gatherer lifeways.

William Andrefsky Jr earned his PhD in Anthropology from Binghamton University. He is currently the Edward R. Murrow distinguished professor Emeritus at Washington State University in Pullman, Washington. His research interests include hunter-gatherer organization, lithic technology, and information transmission.

References

Ackerman

(2011) Microblade assemblages in Southwestern Alaska: An early Holocene adaptation. In: Goebel

Buvit

(eds) From the Yenisei to the Yukon: Interpreting Lithic Assemblage Variability in Late Pleistocene/Early Holocene Beringia. Texas: Texas A&M University Press, pp.255–269.

Adams

(2018) ‘Tis the season: An experimental examination of toolkit variability, late Pleistocene-early Holocene interior Alaska. Unpublished PhD dissertation, Department of Anthropology, Washington State University, Pullman.

Ames

Fuld

Davis

(2010) Dart and arrow points on the Columbia Plateau of Western North America. American Antiquity 74: 287–325.

Bettinger

Winterhalder

McElreath

(2006) A simple model of technological intensification. Journal of Archaeological Science 33: 538–545.

Bever

(2000) Paleoindian lithic technology and landscape use in late Pleistocene Alaska: A study of the mesa complex. Unpublished PhD dissertation, Department of Anthropology, Southern Methodist University, University Park.

Binford

(1979) Organization and formation processes: Looking at curated technologies. Journal of Anthropological Research 35: 255–273.

Bleed

(1986) The optimal design of hunting weapons: Maintainability or reliability. American Antiquity 51: 737–747.

Buchanan

Collard

Hamilton

, et al. (2011) Points and prey: A quantitative test of the hypothesis that prey size influences early Paleoindian projectile point form. Journal of Archaeological Science 38: 852–864.

Cattelain

(1997) Hunting during the upper Paleolithic: Bow, spearthrower, or both? In: Knecht

(ed.) Projectile Technology: Interdisciplinary Contributions to Archaeology. New York: Plenum Press, pp.213–240.

10.

Churchill

(1993) Weapon technology, prey size selection, and hunting methods in modern hunter-gatherers: Implications for hunting in the Palaeolithic and Mesolithic. In: Peterkin

Bricker

Mellars

(eds) Hunting and Animal Exploitation in the Later Palaeolithic and Mesolithic of Eurasia. Washington, DC: Archaeological Papers of the American Anthropological Association, vol. 4, pp.11–24.

11.

Clarkson

(2016) Testing archaeological approaches to determining past projectile delivery systems using ethnographic and experimental data. In: Iovita

Sano

(eds) Multidisciplinary Approaches to the Study of Stone Age Weaponry. Dordrecht: Springer, Vertebrate Paleobiology and Paleonathropology Series, pp.1–13.

12.

Clarkson

Hiscock

Mackay

, et al. (2018). Small, sharp, and standardized: Global convergence in backed-microlith technology. In: O’Brien

Buchanan

Eren

(eds) Convergent Evolution and Stone-Tool Technology. Cambridge: MIT Press, The Vienna Series in Theoretical Biology, pp.175–200.

13.

Clarkson

Shipton

(2015) Teaching ancient technology using “hands-on” learning and experimental archaeology. Ethnoarchaeology 7: 157–172.

14.

Cotter

(1938) The occurrence of flints and extinct animals in pluvial deposits near Clovis, New Mexico: Part V1, report on field season of 1937. Proceedings of the Academy of Natural Sciences of Philadelphia 90: 113–117.

15.

Dixon

(2013) Arrows and Atl Atls: A Guide to the Archaeology of Beringia. Report submitted to the National Park Service. Washington, DC: US National Park Service.

16.

Ellis

(1997) Factors influencing the use of stone projectile tips: An ethnographic perspective. In: Knecht

(ed.) Projectile Technology: Interdisciplinary Contributions to Archaeology. New York: Plenum Press, pp.37–74.

17.

Elston

Brantingham

(2002) Microlithic technology in Northeast Asia: A risk minimizing strategy of the late Pleistocene and early Holocene. In: Elston

Kuhn

(eds) Thinking Small: Global Perspectives on Microlithization. Arlington: Archaeological Papers of the American Anthropological Association 12, pp.103–116.

18.

Fitzhugh

(2001) Risk and invention in human technological evolution. Journal of Anthropological Archaeology 20: 125–167.

19.

Graf

Bigelow

(2011) Human response to climate during the Younger Dryas chronozone in central Alaska. Quaternary International 242: 434–451.

20.

Guthrie

(1983a) Osseous projectile points: Biological considerations affecting raw material selection and design among Paleolithic and Paleoindian peoples. In: Clutton-Brock J and Grigson C (eds) Animals and Archaeology, Vol. 1: Hunters and Their Prey. London: Council for Archaeozoology, BAR International Series 163, pp.273–294.

21.

Guthrie

(1983b) Appendix B: Composite bone-stone tool reproduction and testing. In: Powers WR, Guthrie RD and Hoffecker JF (eds) Dry Creek: Archeology and Paleoecology of a Late Pleistocene Alaskan Hunting Camp. Report submitted to the National Park Service. Washington, DC: US National Park Service, pp.348–374.

22.

Hayden

Franco

Spafford

(1996) Evaluating lithic strategies and design criteria. In: Odell

(ed.) Stone Tools: Theoretical Insights into Human Prehistory. New York: Plenum Press, pp.9–49.

23.

Haynes

(1964) Fluted projectile points: Their age and dispersion. Science (New York, N.Y.) 145: 1408–1413.

24.

Hirasawa Y and Holmes C (2016) The relationship between microblade morphology and production technology in Alaska from the perspective of the Swan Point Site. Quaternary International 442: 1–14.

25.

Hughes

(1998) Getting to the point: Evolutionary change in prehistoric weaponry. Journal of Archaeological Method and Theory 5: 345–408.

26.

Iovita

Sano

(2016) Multidisciplinary Approaches to the Study of Stone Age Weaponry. Dordrecht: Springer.

27.

Knecht

(1997a) Projectile Technology: Interdisciplinary Contributions to Archaeology. New York: Plenum Press.

28.

Knecht

(1997b) Projectile points of bone, antler, and stone: Experimental explorations of manufacture and use. In: Knecht

(ed.) Projectile Technology: Interdisciplinary Contributions to Archaeology. New York: Plenum Press, pp.191–212.

29.

Kunz

Bever

Adkins

(2003) The Mesa Site: Paleoindians Above the Arctic Circle. BLM-Alaska Open File Report 86: Prepared by the Bureau of Land Management.

30.

Lee C and Goebel T (2016) The Slotted Antler Points from Trail Creek Caves, Alaska: New Information on Their Age and Technology. PaleoAmerica 2: 1–8.

31.

Myers

(1989) Reliable and maintainable technological strategies in the Mesolithic of mainland Britain. In Torrence

(ed.) Time, Energy, and Stone Tools. Cambridge: Cambridge University Press, pp.78–91.

32.

Pargeter

(2011) Assessing the macrofracture method for identifying stone age hunting weaponry. Journal of Archaeological Science 38: 2882–2888.

33.

Pétillon

Bignon

Bodu

, et al. (2011) Hard core and cutting edge: Experimental manufacture and use of Magdalenian composite projectile tips. Journal of Archaeological Science 38: 1266–1283.

34.

Pétillon

Plisson

Cattelain

(2016) Thirty years of experimental research on the breakage patterns of stone age osseous points: Overview, methodological problems, and current perspectives. In: Iovita

Sano

(eds) Multidisciplinary Approaches to the Study of Stone Age Weaponry. Dordrecht: Springer, pp.47–63.

35.

Potter

(2011) Late Pleistocene and Early Holocene assemblage variability in Central Alaska. In: Goebel

Buvit

(eds) From the Yenisei to the Yukon: Interpreting Lithic Assemblage Variability in Late Pleistocene/Early Holocene Beringia. College Station: Texas A&M University Press, pp.215–233.

36.

Powers

Guthrie

Hoffecker

(1983) Dry Creek: Archaeology and Paleoecology of a Late Pleistocene Alaskan Hunting Camp. Washington, DC: US National Park Service.

37.

Rasic

(2008) Paleoalaskan adaptive strategies viewed from Northwestern Alaska. Unpublished PhD dissertation, Department of Anthropology, Washington State University, Pullman.

38.

Robinson

(2008) Microblades and seasonality: Settlement patters in the Tangle lakes region of Alaska. The Review of Archaeology 29: 21–40.

39.

Schiffer

(2011) Studying Technological Change: A Behavioral Approach. Salt Lake City: University of Utah Press.

40.

Schiffer

Skibo

(1987) Theory and experiment in the study of technological change. Current Anthropology 28: 595–622.

41.

Shott

(1997) Stones and shafts redux: The metric discrimination of chipped-stone darts and arrow points. American Antiquity 62: 86–101.

42.

Thomas

(1978) Arrowheads and atlatl darts: How the stones got the shaft. American Antiquity 43: 461–472.

43.

Torrence

(1989) Time, Energy, and Stone Tools. Cambridge: Cambridge University Press.

44.

Waber

(2011) The development of the microblade industry at the Richardson Island Site, Haida Gwaii, British Columbia. Unpublished Thesis, Department of Anthropology, University of British Columbia, Vancouver.

45.

Wood

Fitzhugh

(2018) Wound ballistics: The prey specific implications of penetrating trauma injuries from osseous, flakes stone, and composite inset microblade projectiles during the Pleistocene/Holocene transition, Alaska U.S.A. Journal of Archaeological Science 91: 104–117.

46.

Wygal

(2009) Prehistoric colonization of Southcentral Alaska: Human adaptations in a post glacial world. Unpublished PhD dissertation, Department of Anthropology, University of Nevada, Reno.

47.

Wygal

(2010) Prehistoric upland tool production in the Central Alaska range. Alaska Journal of Anthropology 8: 107–119.

48.

Wygal

(2011) The microblade/non-microblade dichotomy: Climatic implications, toolkit variability, and the role of tiny tools in Eastern Beringia. In: Goebel

Buvit

(eds) From the Yenisei to the Yukon: Interpreting Lithic Assemblage Variability in Late Pleistocene/Early Holocene Beringia. College Station: Texas A&M University Press, pp.234–254.

49.

Wygal

(2018) The peopling of Eastern Beringia and its archaeological complexities. Quaternary International 466: 284–298.