Contracting Stemmed: What's the Point?

CONTRACTING STEMMED: WHAT'S THE POINT? Author(s): Robert F. Boszhardt Source: Midcontinental Journal of Archaeology , Spring, 2002, Vol. 27, No. 1 (Spring, 2002), pp. 35-67 Published by: Taylor & Francis, Ltd. on behalf of the Midwest Archaeological Conference, Inc. Stable URL: https://www.jstor.org/stable/20708168 REFERENCES Linked references are available on JSTOR for this article: https://www.jstor.org/stable/20708168?seq=1&cid=pdfreference#references_tab_contents You may need to log in to JSTOR to access the linked references. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms Taylor & Francis, Ltd. and Midwest Archaeological Conference, Inc. are collaborating with JSTOR to digitize, preserve and extend access to Midcontinental Journal of Archaeology This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms CONTRACTING STEMMED: WHAT'S THE POINT? Robert F. Boszhardt ABSTRACT Several authors have argued for an early introduction of bow and arrow technology in eastern North America, frequently citing as evidence a number of relatively small Late Archaic point types. However, these interpretations do not account for the widespread appearance of relatively large, straight- and contracting-stemmed points that mark the transition to the Woodland tradition. Data from the Driftless Area of southwestern Wis consin support an alternative hypothesis regarding the function of small Late Archaic points. It is suggested that the small, notched points served as dart tips that were fixed to detachable wooden foreshafts. These composite points were replaced in Woodland times by larger, straight- and contracting-stemmed forms that served to economically combine foreshaft and tip attributes. Introduction The introduction of bow and arrow technology in North America has been dis cussed in a series of articles over several decades (e.g, Amick 1994; Blitz 1988; Corliss 1980; Odell 1988; Pyszczyk 1999; Seeman 1992; Shott 1993,1997; Tho mas 1978). These have considered the antiquity of the technology and its distri bution, uses, associated social conditions, and means of archaeological identifi cation, including ethnoarchaeological analyses. The latter concern has led to the development of statistical formulas to test whether projectile points can be reli ably interpreted as either arrow or spear/dart tips based on metric attributes. Sev eral authors have challenged the widely held notion that bow and arrow technol ogy was introduced relatively late in prehistory, suggesting that earlier forms of small projectile points may have functioned as arrow tips. Recently, Nassaney and Pyle (1999) addressed the problematic occurrence of small projectile points in cultural contexts that substantially predate the presumed late prehistoric adoption of the bow and arrow in eastern North America. Using metric data from Plum Bayou in central Arkansas, they developed a model to distinguish arrow from dart points, suggesting that length measurements for dart points in that locale tended to be 36 mm and greater while arrow tip lengths were less than 36 mm (1999:250). Nassaney and Pyle further segregated the two classes according to distinct reduction trajectory models reflected in Plum Bayou as semblages: darts were manufactured through bifacial reduction while arrows Midcontinental Journal of Archaeology, Vol. 27, No. 1 ? 2002 by The University of Iowa This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms 36 Robert F. Boszhardt reflect flake retouching. They also provided a valuable pan-regional comparison of dart and arrow typologies, revealing the widespread occurrence of small "ar row-size" points in earlier traditions. While some of these small points extend back to the Paleoindian tradition (e.g, Amick 1994), most correspond to Late Archaic cultures (Nassaney and Pyle 1999:258). Assuming that small points rep resent arrow tips, Nassaney and Pyle concurred with Bradbury (1997) and others that the bow and arrow likely appeared periodically during earlier stages of pre history. An alternative hypothesis for at least some small Late Archaic projectile point types is that they represent dart points that would have been affixed to com pound shafts. Key to this hypothesis is examining the questions of when and why people manufactured relatively large, contracting-stemmed projectile points. These bifacially flaked points are known by a variety of names, including Adena, Belknap, Dickson, Gary, and Waubesa, and occur throughout much of eastern North America (Justice 1987:189). While contracting-stemmed points may have persisted in some regions longer than in others, they appear to have been most common in the Midwest between ca. 2500 and 1800 B.P. (Emerson 1986:624) and in some regions replaced small "arrow-sized" points such as Merom Ex panding Stemmed and Trimble Side Notched (Justice 1987:130-132; Winters 1969). This period coincides with the transition from Archaic to Woodland in the Midwest. Driftless Area Point Sequence Point data for the type sequence spanning the Archaic-Woodland transition in the Driftless Area of southwestern Wisconsin provide a basis for examining tech nological adjustments that led to the adoption of the Waubesa contracting-stemmed point type. The Driftless Area, a rugged landscape within the Upper Mississippi River basin, was missed by glacial advances (Martin 1965). Mid-twentieth-cen tury excavations of stratified rockshelters such as Raddatz (Wittry 1959a) and Durst (Wittry 1959b) provided a basic projectile point chronology for the region. These baseline studies have been augmented by subsequent research, including intensive investigations in the Kickapoo River valley (e.g., Halsey 1976) and at a series of Driftless Area sites examined during the career of James Stoltman at the University of Wisconsin-Madison (e.g., Stoltman 1998). The regional point sequence begins with Early Paleoindian fluted points, fol lowed by Late Paleoindian lanceolate types. The transition to the Archaic tradi tion is recognized by the earliest notched and stemmed forms such as St. Charles, Thebes, and Hardin Barbed. The majority of these early points are relatively large, although diminutive examples of fluted points are known (Amick et al. 1999). For purposes of this study, common local types representing the sequence from the Middle-Late Archaic through Early Woodland stages will be consid ered. These begin with Raddatz side-notched, followed by Preston corner-notched and Durst expanding-stemmed, followed in turn by straight-stemmed Kramer and contracting-stemmed Waubesa points. While these types are regional, the This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms Contracting-Stemmed Points 37 forms represented in this sequence are found in a much broader geographical area. For comparative purposes, a local Early Archaic type (Price stemmed) and late prehistoric "arrow" point types (Honey Creek corner-notched, Cahokia/Grant side-notched, and Madison triangular) are also considered (Figure 1). Price stemmed points are relatively common in western Wisconsin but have had a problematical taxonomic history. In general, these are relatively thick and crudely flaked lanceolate bifaces with slightly expanding stems. Examples were recovered in the Kickapoo River valley and initially interpreted as Late Paleoindian in age (Hurley 1965). However, Freeman (1966:41-43) interpreted the Price Stemmed type as Early-Middle Woodland, and the examples from the Kickapoo River valley were subsequently reinterpreted as representing these later cultures. More recently, Stoltman has adopted the Matanzas name for flat-based examples, associating them with the 5,000-6,000-year-old Helton phase in Illi nois (1998:123). Twelve Price stemmed points were recovered beneath levels containing Late Archaic Durst and Preston corner-notched points in the lowest levels of the Lawrence Rocksheiter (47Vel54) in the Kickapoo Valley (Halsey 1976). Some of these points have distinctly concave bases and are comparable with the Early Archaic Dalton/Quad cluster (Justice 1987:35-36). At the Raddatz and Durst Rockshelters, the Middle Archaic levels were domi nated by Raddatz side-notched points. This type compares to large side-notched points from adjacent regions, such as Big Sandy, Godar, Hemphill, Matanzas, and Osceola (Ahler 1993:485; Justice 1987:60-71). These points are tradition ally considered spear or dart tips (at least one of the rocksheiter specimens ex hibits an impact fracture), and nearly all specimens fall within the size range of Nassaney and Pyle's dart grouping. Raddatz points have straight edges and some from the type site were clearly retouched, indicating multiple use as hafted knives. The Raddatz levels in the rockshelters were stratigraphically overlain by compo nents with expanding-stemmed Durst points, which conform to the Lamoka cluster (Justice 1987:127). Raddatz components are well dated to ca. 5000-3000 B.P., while Durst points mark the local Late Archaic stage and are dated from ca. 3000 to 2500 B.P. (Stoltman 1998). Stoltman (1998:134) has recognized Preston corner-notched as a small Late Archaic point variant in the Driftless Area. Examples of this type were found beneath Durst layers at the Preston Rocksheiter in Grant County, Wisconsin, and Stoltman estimates the age of the Preston phase as 3500-3000 B.P. Both Preston and Durst points have expanding stems, and Stoltman describes the Preston type as "a small (ca. 4.5 cm=< 2 in. long), diagonal-to-side-notched projectile point" (1998:134). This size is within the range for dart tips, according to Nassaney and Pyle's groupings; however, all but one of the Preston points in the type set illus trated by Stoltman are less than 40 mm long (1998:Figure 6.14). The type set includes former "Monona Stemmed" specimens from the Durst Rocksheiter, with lengths ranging from 26 to 38 mm, and averaging 35.8 mm. Furthermore, Stoltman compared the Preston type to Merom and Trimble points of the Riverton culture in the Wabash River valley (Winters 1969), and Nassaney and Pyle consider Merom and Trimble as possible Archaic arrow tips (1999: 256). This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms 38 Robert F. Boszhardt c 1 ? < I ? ^^^^^^^^^^^ ^^^^^^ 3 ^^^^^^^ ^^^^^^^^^ ? > ^^^B^^ * 'S ?9 I i fi ?s '? 1 This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms Contracting-Stemmed Points 39 Following the Preston and Durst phases, projectile points of the Driftless Area included relatively large, straight-stemmed Kramer points that were soon re placed by contracting-stemmed Waubesa points. Coupled with the first evidence for ceramics in this region, these point types are local markers for the onset of the Woodland tradition (Stevenson et al. 1998). In the Driftless Area, Kramer points are diagnostic of the Indian Isle (formerly Ryan) phase (ca. 2500 B.P.), while Waubesa points are affiliated with the Prairie phase (ca. 2000 B.P.), in cluding direct and exclusive association in sealed floodplain contexts (Stoltman 1990; Theler 1987) Unfortunately, published metric data are generally lacking for these types in this region, largely because contracting-stemmed points are, for some reason, extremely rare in local rocksheiter sites. Nonetheless, Hurley (1965:Figure 3) illustrated a typical collection from the Kickapoo River valley that shows larger-sized Raddatz and Waubesa points in comparison to Durst points in that study area. Prevailing interpretations of the introduction of bow and arrow technology in this region are based on the relatively widespread occurrence of small notched (e.g., Honey Creek corner-notched and Cahokia and Grant side-notched) and unnotched (Madison triangular) projectile points beginning around A.D. 500 700 (Benn 1979; Mead 1979; Stevenson et al. 1998). These point types are ubiq uitous at Late Woodland and Oneota sites in this region. While most are bifa cially flaked (presumably from flake blanks), some were manufactured by sim ply trimming flake edges into shape. Individual point weights for these arrow tips are usually 1-2 g. In sum, the Middle Archaic to Early Woodland point sequence for the Driftless Area begins with the relatively large Raddatz side-notched type, shifts to rela tively small Preston and Durst expanding-stemmed forms, and culminates in relatively large Kramer and Waubesa points. In the scenario outlined by Nassaney and Pyle, Raddatz points would be classified as dart points, many Preston and Durst points would be considered arrow tips, and Kramer and Waubesa points would fall into the dart category. However, published data indicate that, while Preston and Durst points tend to be notably smaller and lighter than preceding Raddatz and succeeding Kramer and Waubesa points, the smaller Late Archaic types are substantially larger and heavier than late prehistoric arrow tips. Table 1 presents metric point data compiled from published sources on se lected Driftless Area components. This compilation provides insights into broad patterns but incorporates inevitable comparative unknowns because of unstandardized measuring methods used by various researchers. In order to cor rect for any such biases, 385 Driftless Area points representing nine types were measured using a standard protocol. Remeasurement was undertaken of the Raddatz and Durst Rocksheiter assemblages, and material from several compo nents not included in Table 1 was examined. Among the latter were points from the large Early Woodland assemblage from the Beach site near Madison (Salkin 1986), the stratified Preston Rocksheiter in Grant County, and Gullickson's Glen Rocksheiter in Jackson County, Wisconsin. Preston Rocksheiter was excavated by Harris Palmer and James Stoltman, revealing a long sequence of occupation, This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms 40 RobertF Boszhardt < <N ^ O < - < < - on ^ on. < < oi >n oo <^ (Ni w |> ^ 00 < 00 C ? ig 00. ^C?n I ( CT $ < Q co ? ? m 3 rn _ <?> !Q 0\ w ? ^ ^ < q q O < co ^- ' < on -3- w A q ^ oo? ro co ^ ? on ? -h' oo ?o _j m i- ^ 5\ h m - wO co w co w m cN w o < o ? 2 < ? ( < CO co t? co > t? o q vO wo ^ ? ? 'S ? 's <? J Cu ,? CO ?j o ?i o- o o vd co t? co ^ .a 2 S ?1 .2 o .a o? Q te 0? a -a w 00 t? te .2 'S Q O On r?t o i- CO t? q < ? g 3^ c g 2 b -S JS o11 ? S-l? S S -g ? Cu w c ? < This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms 1 td t3 ?n ?4 co t? ?4 - J2 -a -a 3 t? G Contracting-Stemmed Points 41 and the materials are curated at the University of Wisconsin-Madison, as are the Beach site materials. Gullickson's Glen was excavated by Warren Wittry, who recovered primarily Woodland and Oneota materials, which are curated at the Wisconsin Historical Society, as are the Durst and Raddatz assemblages. At tributes were measured using dial metric calipers and a portable digital scale and included length, haft width, weight, shoulder width, and maximum thickness. Obviously, breakage precluded accurate measurement of attributes such as length and weight for some specimens. In some assemblages with numerous examples of specific types (e.g., Waubesa points from the Beach site and Madison triangu lar points from Gullickson's Glen), only complete points were included in the tabulations. Note was made of points that appeared to have been resharpened, but such points were not excluded from summary calculations. These data are presented in Tables 2-10. Table 11 presents the composite data from the 385 measured points, and these data are illustrated in Figure 2. In general, size patterning in the evolution of Archaic to Early Woodland point types in this region is comparable in the two samples summarized in Tables 1 and 11. Raddatz points tend to be larger than Preston and Durst types, and some Preston and Durst points are so diminutive as to raise the question of whether or not they represent arrow tips. The published lengths of 11 measurable Raddatz points from the type site range from 38 to 64 mm, with an average of 49.4 mm (Table 1), and Wittry (1959a:46) considered this to be a minimum length due to common evidence of resharpening. The published length range of Durst points from the nearby Durst Rocksheiter is 21.5-42.0 mm (average 30.78 mm). Direct measurements of 40 Raddatz and 100 Durst points show a similar average length difference (46.39 and 37.50 mm, respectively), and the average for 33 Preston points (33.44 mm) is smaller than that for Durst. This size difference is clarified in weight comparisons of these point types: Raddatz averages 9.44 g, Preston 4.22 g, and Durst 4.65 g. Consequently, on the basis of length, many of the Preston and Durst points conform to Nassaney and Pyle's arrow tip class. Points of the Kramer and Waubesa types, representing the transition to the Woodland tradition, are substantially larger than the Preston and Durst points of the preceding terminal Archaic stage. Measurement of 16 Kramer and 30 Waubesa points reveals a notable increase in size (Table 11). For example, the Early Wood land styles average nearly 20 mm longer and are over twice as heavy as their Late Archaic predecessors (10.20 and 13.21 g for Kramer and Waubesa, respec tively, compared with 4.22 and 4.65 g, respectively, for Preston and Durst). The only attribute that does not reflect this pattern so dramatically is shoulder width: Preston and Durst points are only slightly narrower than Raddatz and Kramer/ Waubesa points. Kramer and Waubesa points compare with Means and Gary types described by Nassaney and Pyle (1999) for the Plum Bayou area and can be classified with confidence as dart points. Why Contracting Stems? A significant aspect of the shift to Kramer and Waubesa points is that all of the This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms 42 Robert F. Boszhardt Table 2. Composite Measurements of Early Archaic Price Stemmed Points. Haft Shoulder Specimen* 6A-3(4)1 4A-3(4)1 WPS-72 6A-4(1)1 6C-5(6)1 6C-6(2)1 6C-5(6)1 5C-3(5)1 Length Width Weight Width Thickness (mm) (mm) (g) (mm) (mm) 18.5 19.8 54.0 61.5 50.4 17.4 19.6 16.2 19.7 17.5 18.5 15.2 18.5 26.9 21.1 17.4 63.0 49.5 WPS-11 1C-4(1)1 6F-5-14 50-4(2)1 Average 44.5 41.5 52.1 14.1 15.5 11.3(+) 17.7 11.8 8.3 10.9 21.2 23.4 24.1 22.7 25.6 22.6 30.0 19.5 28.2 21.5 24.5 24.0 9.5 10.6 10.2 9.5 9.2 10.0 9.6 12.0 9.2 8.6 9.4 7.5 9.6 Comments impact fracture, tip gone (N=12) *Bard Lawrence Rocksheiter preceding Archaic forms are either distinctly notched or have expanding stems. These bases allow fixed hafting, so removal would have been possible only by cutting or untying the lashing. Raddatz, Preston, and Durst points therefore were apparently manufactured and hafted with the intent that they stay on the shaft during use. In contrast, the subsequent Early Woodland straight-stemmed and, especially, the contracting-stemmed points were presumably designed to facili tate removal from the haft?but why? What technological advantages did these more readily detachable points provide that led to their widespread adoption at the transition from the Archaic to the Woodland tradition? Consideration of how straight- and contracting-stemmed points might have been hafted is initially confounding. Webb's early description of contracting stemmed points from Indian Knoll in Kentucky includes this observation: "One would suspect that [this form] was particularly difficult to attach firmly to the shaft. Perhaps it was intended to be easily detached" (1946:254). Shortly there after, he described similar points from the Carlson Annis Mound in Kentucky: "This type is a large heavy blade with a short crude stem. The stem is thick at the base and expands from the base to form very weak shoulders. It is difficult to see how such a stem was of any value is [sic] hafting the projectile" (Webb 1950:307). Apparently, the shaft end was split or notched, the point stem inserted (per haps affixed with a mastic such as pitch), and then clamped in place by lashing the shaft end with sinew or other fibrous cordage. Because these points lack notching or grinding on the blade edges, the lashing must have been restricted to the haft end, encompassing only the stemmed portion of the point, which, as Webb observed, is often quite short. Unfixed point tips offer their users several advantages, for example, remain ing in wounds to cause further bleeding. This particular feature may be evident This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms Contracting-Stemmed Points 43 Table 3. Composite Measurements of Mid-Late Archaic Raddatz Side Notched Points. Haft Width Specimen* Length (mm) [R] Ell-6-1 58.7(+) 18.7(+) 54.2(+) 53.5 45.2 19.2 19.5 21.4 21.2 19.0 38.2 52.8 15.8 17.9 5.5 10.4 [R] El 1-6-2 [R] D12-9-1 [R] D12-6-2 [R] D12-6-1 [R] D12-6-3 [R]E10-10-1 [R] D10-6-4 [R]D10-6-l [R] D10-6-2 [R] D9-10-1 [R] D9-8-1 [R]E12-5-l (mm) 17.1 17.5 17.1 18.5 18.1 49.6 [B] 104-6 67.5 19.2 19.0 21.2 21.4 17.4 21.0 24.5 26.2 [B] 6-42 77.5 23.5 [R]E12-6-2 [R] 6.2 57.0(+) 42.0 [R] 6.1/7.1 [R] H12-7-1 [R] 12.1 69.8 17.5 14.9 25.4 33.5 19.3(+) 13.9 7.6 11.9 17.2(+) 8.8 20.5 20.8 [B] 47-15 [B] 119-3 [B] 21-36 23.1 [B] 3-26 18.5 Shoulder Width Thickness (mm) (mm) 28.4 9.3 28.7 10.5 27.5 11.5 26.4 10.0 26.6 7.2 22.5 9.5 21.4 8.2 26.3 8.9 7.8 26.7 26.2 27.9 7.2 9.1 24.8 9.1 27.2 8.9 26.0 9.5 29.1 9.4 28.0 8.1 23.0 6.5 27.7 9.8 34.5 8.4 9.4 33.4 20.6 [B] 68-8 [B] 27-17 (8) 20.9 [R] E12-6-1 [R] F-7-1 Weight 21.6 12.5 27.6 8.6 27.5 12.0 27.5 28.8 27.5 10.0 8.1 8.9 24.0 11.5 [B] 107-9 28.0 17.1 3.9 21.6 7.4 [B] 9-52 42.8 13.5 6.1 18.6 8.4 [B] 7-11 52.9 22.0 15.3 29.5 9.9 [B] 47-16 56.3 17.4 12.0 24.4 10.3 [B] 122-12 50.0 16.5 8.5 23.7 8.2 [B] 13.22 44.0 20.0 6.7 25.5 7.0 [B] 26-6 55.4 25.7 13.8 29.4 9.5 Comments extreme tip missing concave base; tip missing base wider than shoulder base wider than shoulder impact fracture; tip gone base wider than shoulder tip missing base only extreme tip missing tip missing concave base concave base; base wider than shoulder concave base; base wider than shoulder concave base; base & shoulder same width concave base; shoulder wider than base concave base extreme concave base concave base; base wider than shoulder concave base; base wider than shoulder concave base; small specimen straight base; base wider than shoulder straight base; base wider than shoulder straight base; shoulder wider than base convex base; base wider than shoulder straight base; base wider than shoulder straight base; base wider than shoulder This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms 44 Robert F. Boszhardt Table 3. Continued. Haft Shoulder Specimen* Length Width Weight Width Thickness (mm) (mm) (g) (mm) (mm) [B] 31-13 46.4 18.6 9.4 24.0 8.4 [B] 62-21 [B] 111-31 40.4 40.6 19.5 8.5 26.4 6.5 9.8 [B] 68-15 35.1 15.9 4.6 23.7 7.2 [BL]10-4(1E)2 46.4 Average (N= 40) 16.1 9.4 20.6 26.5 7.3 9.0 19.5 Comments straight base; shoulder wider than base broken straight base; base wider than shoulder straight base; base wider than shoulder *[B] Beach site (only whole points measured) [BL] Bard Lawrence Rocksheiter [R] Raddatz Rocksheiter in the Indian Knoll cemetery, where Webb reported that the contracting-stemmed form "is interesting because, while it is less than one percent of the total, it ap pears often in burial association in such situations as to suggest that instead of being a burial offering, such specimens may have been imbedded in the flesh of the body at burial and may have been the cause of death" (1946:255; see also Webb 1946:224-246 and Figure 29D a, c, showing contracting-stemmed and straight-stemmed points embedded in human vertebrae). In addition, detachable point tips allow recovery of the shaft, which, accord ing to modern dart replicators, requires more construction time and effort than point manufacture (Loren Cade, personal communication). In effect, the straight stemmed Kramer and contracting-stemmed Waubesa points replicate compound dart shaft technology, with the added intent of having the tip separate from the shaft after penetrating the target. Indeed, the wider shoulders of some straight and contracting-stemmed point blades create barb-like projections that would facilitate removal of the point from the shaft. Such projections are emphasized in a series of Late and Terminal Archaic straight-stemmed points in the central and lower Mississippi Valley such as Etley barbed, Delhi, Wade, and Buck Creek barbed (Justice 1987:146,179), and are more pronounced yet on Eva points (Jus tice 1987:100-103). Contracting-Stemmed Knives? Based on the stem form and lack of notching, contracting-stemmed bifaces would appear to be ill suited for hafting and for use as knives or other tools where lateral pressure to the blade is involved. However, Price (1986:539) interpreted contracting-stemmed bifaces from the Early Woodland Tchula horizon in the Little Black River area of the Ozarks as having been resharpened from use as knives. Similarly, Emerson (1983) and Emerson and Fortier (1986) suggested This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms Contracting-Stemmed Points 45 Table 4. Composite Measurements of Late Archaic Preston Corner Notched Points. Specimen* D] F5-4-1 D] C5-3-2 D] F5-4-2 D] D] D] D] F4-6-1 E5-4-1 C5-4-1 D4-4-1 Haft Length (mm) Width 30.5 38.5 37.3 33.6 25.6 10.9 9.5 10.7 10.0 10.0 11.9 10.9 10.2 9.6 9.3 9.5 14.5 11.5 D] F4-9-2 D] F6-7-1 D] E5-4-2 D] C5-3-3 BL] 5B-3-1 BL] 5A-3(5)1 BL] 4C-3(2)3 BL] 3-F.113-1 P]416 P] 628 P] 627 P] 395 P] 237 P] 696 P] 646 P] 350 P] 647 P] 349 P] 623 P] 397 P] 384 P] 282 P] 391 P] 245 P] 165 P] 409 Average (N=33) 33.5 39.2 28.0 25.6 37.0 41.2 30.0 31.7 41.5 37.0 28.2 30.0 33.4 (mm) 15.1 11.5 14.1 12.5 12.6 11.0 10.9 14.1 13.0 14.5 9.4 13.0 12.0 12.1 12.6 15.0 12.2 11.0 12.2 13.5 11.8 Shoulder Weight Width (g) (mm) 3.2 3.9 5.3 4.1 4.1 5.2 4.2 4.3 3.0 5.5 6.0 3.7 2.8 5.1 5.4 2.4 3.6 Thickness (mm) 23.4 18.0 23.0 24.6 23.5 23.3 16.4 7.1 7.3 8.5 7.9 7.9 17.5 6.8 5.4 6.2 4.2 5.8 6.5 6.5 7.0 6.8 6.5 7.9 6.6 5.1 6.2 9.4 7.4 8.0 6.0 27.0 22.8 9.4 8.0 18.5 18.9 21.1 21.8 21.8 21.9 21.5 19.0 19.0 18.5 16.5 27.0 22.8 24.5 19.6 25.7 20.0 26.7 4.2 25.9 21.9 Comments tip crushed impact fracture 6.3 7.2 7.6 7.3 *[D] Durst Rocksheiter [BL] Bard Lawrence Rocksheiter [ ] Preston Rocksheiter that a set of "Goose Lake knives," bifaces with short contracting stems from the Florence Street site in Illinois, functioned as cutting tools. Only three of the 36 contracting-stemmed bifaces from Florence Street were classified as Goose Lake knives, based in part on relatively fine flaking and observed, though undescribed, microscopic use wear. Farnsworth and Asch (1986:366) noted the similarities between Goose Lake knives and Peisker Diamonds, as well as their restricted distributions to the American Bottom and Lower Illinois valley, respectively. A This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms 46 Robert F. Boszhardt Table 5. Composite Measurements of Late Archaic Durst Expanding Stemmed Points(all chert unless otherwise noted). Haft Shoulder Thickness Specimen* [D] G4-5-3 [D] F6-6-1 [D] B6-1-4 [D] E4-5-1 [D] L6-3-1 [D] F7-3-1 [D] C7-4-2 [D] F6-6-3 [D] C8-4-2 [D] F5-5-15 [D] D6-4-1 [D] F6-6-2 [D] G4-5-1 [D] D7-5-3 [D] E5-2-1 [D] G4-3-1 [D] C7-4-1 [D] F5-6-2 [D] D8-5-2 [D] E6-2-1 [D] D5-5-2 [D] E6-5-1 [D] F4-6-2 [D] F4-5-1 [D] E8-5-1 [D] F7-4-3 [D] F4-4-1 [D] G4-5-2 [D] D5-4-2 [D] F7-5-1 [D] F4-6-3 [D] E6-6-1 [D] C7-2-10 [D] F4-5-2 [D] E4-3-1 [D] D7-5-2 [D] D4-5-5 [D] D8-5-3 [D] F4-5-4 [D] F6-6-5 [D] D8-5-4 [D] G4-4-1 [D] D7-4-2 [D] F7-4-2 [D] G4-1-1? [D] G4-3-2 [D] F6-6-6 [D] B7-2-3 Length Width Width Maximum Weight (mm) (mm) (mm) (mm) (g) 29.0 35.0 34.0 42.0 31.0 35.5 39.8 34.0 30.5 31.0 27.0 35.4 27.8 29.5 23.2 28.5 35.0 29.0 31.5 27.3 11.0 11.0 11.0 12.0 11.0 12.0 9.0 9.5 10.5 11.5 10.2 10.2 10.5 11.0 10.0 11.5 11.5 9.0 7.5 11.5 11.0 10.0 8.0 11.1 11.0 8.5 10.3 10.2 10.0 11.0 11.6 11.3 13.2 10.0 12.0 14.3 10.3 10.2 11.7 12.2 12.5 17.5 17.9 18.2 17.6 19.2 20.6 16.2 14.1 14.9 21.8 17.1 16.1 14.8 17.8 16.1 16.9 14.5 14.1 18.4 15.0 16.5 17.1 21.0 19.5 20.5 6.6 7.0 7.0 10.1 8.5 7.2 6.9 6.0 7.8 8.5 7.7 6.9 6.9 7.2 7.5 7.0 7.0 7.0 5.0 9.0 6.2 5.8 6.9 7.2 3.2 2.9 2.7 3.9 2.5 3.2 2.0 1.7 5.1 shoulder snap 2.7 2.6 2.5 6.0 6.9 6.1 7.7 17.9 25.2 15.4 7.7 18.5 18.0 3.7 5.8 basal snap basal snap basal snap 6.4 13.8 15.8 18.2 21.1 28.7 21.0 16.1 Comments 4.2 3.4 3.7 6.1 4.3 7.3 7.0 6.7 7.1 7.8 impact fracture 7.0 7.0 6.0 7.4 7.8 9.4 10.4 9.2 9.1 10.6 11.6 11.8 This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms base base base base base base base base only only only only only only only only Contracting-Stemmed Points 47 Table 5. Continued. Specimen* [D] [D] [D] [D] Haft Shoulder Thickness Length Width Width Maximum Weight (mm) (mm) (mm) (mm) (g) 11.4 11.4 D6-1-5 G6-4-2 F4-8-2 F7-5-2 base base base base base base 11.1 9.3 10.2 11.6 [D] F7-7-1 [D] D7-5-1 [D] F6-4-1 base only questionably a [BL]5A-3(2)1 60.2 [BL] 50-3(1)2 63.0 40.4 40.1 [BL] 5A-3-F.105-1 41.2 [BL] 70-3(1)1 [BL] 6A-28(1)1 [BL] 6A-3(2)1 [BL] 60-3(1)2 [BL] 50-3(1)1 [BL] 10-3-F-13 [BL] 2A-2(?)2 [BL] 6A-3F13-4 [BL] 4A-3(1)1 [BL] 3A-3(2)3 [BL] 5C-3(2)1 [BL] 3C-3(2)1 [BL] WPS-1 [BL] 3C-3(3)1 [P] 561 [P] 176 [P] 360 [P] 1?9 [P] 182 [P] 181 [P] 620 [P]314 [P] [P] [P] [P] [P] [P] [P] 448 346 140 159 540 375 406 12.9 8.2 36.2 41.4 35.5 41.2 38.0 40.5 35.5 31.4 27.1 28.3 28.3 25.5 27.7 81.0 61.6 42.4 34.6 36.0 47.5 45.2 37.0 61.8 44.3 37.6 20.4 7.9 8.6 10.5 [BLJ4B-3-1 [BL] 6A-2B(1)2 only only only only only only quartz; base only [D] C7-2-9 [DJC7-2-3 [BL] 3A-3(2)1 Comments 13.6 16.6 12.0 13.0 9.4 12.5 9.5 9.6 8.4 6.1 7.6 12.8 10.5 10.0 10.5 3.4 1.9 10.6 12.1 7.0 16.0 12.5 12.3 11.1 11.1 10.0 10.0 9.6 11.5 11.2 11.0 13.5 12.8 12.3 10.4 5.5 5.0 7.0 4.3 5.2 4.3 3.6 3.8 4.4 2.3 1.9 1.6 17.3 10.6 5.8 3.7 3.8 7.4 5.2 3.4 9.4 6.4 4.6 "Durst" point serrated edge, unfin ished? sil. sandstone serrated edge, 21.9 6.9 23.3 25.4 21.8 21.4 18.5 20.0 16.5 17.9 14.6 8.0 7.4 7.5 6.9 10.0 20.0 18.6 16.2 14.2 8.5 6.6 4.9 6.5 7.1 6.5 13.4 13.6 6.3 5.2 7.9 10.2 7.6 6.8 6.9 8.4 20.0 21.4 20.6 25.0 6.6 8.6 7.8 7.8 8.0 5.9 6.5 25.8 11.5 19.1 silicified sandstone, tip gone 6.9 10.0 7.9 17.4 32.2 22.8 19.0 19.0 18.5 22.0 19.7 17.5 unfinished? 7.0 This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms silicified sandstone 48 Robert F. Boszhardt Table 5. Continued. Haft Shoulder Thickness Length Width Width Maximum Weight (mm) (mm) (mm) (mm) (g) Specimen* [P] 413 [P] 539 [B] 35-14 [B] 125-3 [R] D12-3-2 37.0 33.0 37.9 [R] D12-3-1 [R] 2.1 [B] 125-3 Average (N=100) 33.0 37.5 10.0 11.0 10.8 10.5 11.5 11.0 11.0 11.0 3.6 2.8 4.1 2.8 12.0 Comments 17.0 16.4 18.8 23.1 18.8 6.2 7.0 6.4 6.2 7.8 16.4 13.2 7.0 6.1 silicified sandstone; tip missing *[D] Durst Rocksheiter [BL] Bard Lawrence Rocksheiter (whole points only) [P] Preston Rocksheiter [R] Raddatz Rocksheiter [ ] Beach site (whole points only) Table 6. Composite Measurements of Early Woodland Kramer Straight-Stemmed Points (all chert). Haft Shoulder Specimen* Length Width Weight Width Thickness (mm) (mm) (g) (mm) (mm) 60.4 51.9 18.5 12.7 [B] [B] [B] [B] [B] 72-16 14-18 6-43 21-20 47-10 [P] [P] [P] [P] 253 284 236 95 50.0 34.0 89.1 53.4 60.0 54.6 49.5 [R] D9-10-2 49.8 [P]641 [R]Ell-4-l [R] E12-4-1 [R] F9-6-1 [R] 5.2 [R] H9-7-1 Average (N=16) 61.2 16.4 12.5 43.5 45.9 18.0 18.5 16.9 14.0 15.6 19.4 19.7 15.5 18.9 19.0 17.5 54.1 16.8 16.2 12.4 9.6 9.4 7.6 5.8 20.5 8.9 13.4 8.3 11.7 9.2 7.2 8.6 10.2 27.2 20.1 20.0 20.0 24.4 27.9 21.1 17.7 18.1 22.8 25.0 22.0 20.7 22.0 21.2 22.0 Comments 10.0 10.0 7.5 7.5 17.5 7.4 8.0 9.5 8.9 10.2 9.1 8.0 7.7 8.0 9.8 9.3 * [B] Beach site (whole points only) [P] Preston Rocksheiter [R] Raddatz Rocksheiter This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms tip reworked Contracting-Stemmed Points 49 Table 7. Composite Measurements of Early Woodland Waubesa Contracting Stemmed points (all chert unless otherwise noted). Haft Shoulder Specimen* [B] 8-6 [B] 18-9 [B] 73-4 [B] 6-41 [B] 46-18 Length Width Weight Width Thickness (mm) (mm) (g) (mm) (mm) [B] 67-8 [B] 29-12 [B] 34-15 [B] 6-22 [B] 23-11 [D] E6-2-3 [D] C8-2-2 [P] 379 58.4 57.4 58.5 70.9 53.2 58.5 50.2 61.5 38.0 76.5 55.5 45.5 42.6 38.5 41.0 79.4 57.7 54.6 60.2 45.0 58.3 [MC] 471x602 [MC] 471x602 [GG] II-D-11 52.9 56.1 69 [B] 74-25/8 [B] [B] [B] [B] [B] 37-12 29-18 111-4 9-33 5-13 [B] 29-11 [B] 62-11 [B] 57-11 [ ] 446A 82.0 19.5 15.6 21.0 20.2 14.7 24.5 17.6 19.5 14.2 14.6 19.2 16.8 17.5 17.1 12.3 11.6 18.4 15.4 21.0 14.3 14.5 21.2 14.2 15.8 21.5 32.5 23.6 36.1 29.0 24.1 25.6 28.5 30.2 21.3 22.1 37.4 26.5 24.1 26.7 21.6 16.5 33.0 24.7 9.0 11.0 10.2 8.9 12.3 5.6 13.5 29.0 17.9 28.0 9.5 8.0 10.6 8.8 9.5 14.8 21.4 22.5 28.5 13.3 17.0 11.5 14.4 11.7 11.9 12.8 12.9 6.4 25.8 12.5 6.3 7.7 6.2 4.9 24.1 12.3 15.4 29.5 36.7 Comments 11.5 12.2 10.6 12.5 9.0 11.4 9.5 8.2 7.8 9.3 8.7 rhyolite 11.6 11.8 split longitudinally 10.5 8.9 7.8 8.0 Prairie du Chien chert; heat treated; very [GG] D7-UL-40 82.5 24.0 [GG] C3/C4-27 19.6 [WS] 471x693 58.1 21.6 18.5 17.8 [WS] 471x693 Average (N=30) 57.9 18.7 39.5 9.0 8.2 16.0 2.9 13.2 26.3 9.9 *[B] Beach site [D] Durst Rocksheiter [ ] Preston Rocksheiter [MC] Mormon Coulee [GG] Gullickson's Glen Rocksheiter [WS] West Salem This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms weak shoulders silicified sandstone; weak shoulders Prairie du Chien chert; base only silicified sandstone silicified sandstone 50 Robert F. Boszhardt Table 8. Composite Measurements of Late Woodland Honey Creek Corner Notched points (all chert unless otherwise noted). Haft Shoulder Length Width Weight Width Thickness Specimen* (mm) (mm) (g) (mm) (mm) Comments [D] C7-1-1 [D] F5-1-1 [D] D6-1-4 [D] C6-1-5 [D] B5-1-2 [D] [D] [D] [D] D8-3-1 D6-2-1 D5-3-1 F6-4-3 [D] D5-1-3 [D] Surface [D] D7-2-1 [GG] B-5-U-1 [GG] D7-LL-1 [GG] B3/B4-L-7 [GG] C-5-UL-2 [GG] II-D-6 [GG] C3/C4-18 [GG] B7-L-1 20.5 24.3 20.5 23.6 18.4 22.2 10.2 22.0 30.0 21.0 17.5 19.0 18.5 28.6 19.2 16.5 13.4 10.2 10.2 13.6 9.2 8.4 10.0 10.0 9.0 11.5 13.0 8.0 12.5 14.5 14.5 11.0 10.1 1.3 1.6 1.1 13.2 18 15.6 13.2 15.3 14.5 16.9 12.4 15.4 18.7 12.5 13.5 1.1 14.1 1.1 1.2 1.0 1.3 2.7 1.0 1.4 1.4 [GG] B6/C6-L-21 22.7 Average (N=20) 20.9 11.1 15.7 1.3 4.5 4.1 4.4 serrated 4.6 3.6 3.8 3.5 4.0 3.9 3.2 3.0 5.0 4.5 4.4 5.0 4.5 3.0 3.0 3.6 flake point 13.7 4.0 missing part of corner Prairie du Chien chert; 15.0 4.0 17.0 17.0 13.5 13.8 serrated serrated serrated silicified silicified silicified silicified silicified silicified sandstone sandstone sandstone sandstone sandstone sandstone silicified sandstone; missing part of corner *[D] Durst Rocksheiter [GG] Gullickson's Glen Rocksheiter few straight- and contracting-stemmed bifaces were clearly worked into nar row-bladed tools such as awls or drills (e.g. Conrad 1986:Figure 15.17; Conrad et al. 1986:Figure 10.10; Emerson and Fortier 1986:Figurel 8.14) or hafted scrap ers (e.g., Farnsworth and Ash 1986:Figure 16.17). Awls, drills, and scrapers, like projectile points, generally function through application of force directly from the haft to the tip of the tool. As long as the pressure is applied in a forward direction, the tip is secure in the haft. Theler (1986,1987) also recovered a series of contracting-stemmed bifaces from the sealed Early Woodland component at the Mill Pond site in the flood plain of the Upper Mississippi River near Prairie du Chien, Wisconsin. These were classified as Waubesa points, and Theler's (1986:146) microscopic ex amination of these specimens found no wear. The Mill Pond assemblage in cluded both complete specimens and several with transverse lateral fractures. Of interest is that the fractures on the Mill Pond contracting-stemmed bifaces occur on the midsections of the blades (Figure 3). The midsection is the thick est portion of the Mill Pond bifaces and seemingly the most difficult to break This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms Contracting-Stemmed Points 51 Table 9. Composite Measurements of Terminal Late Woodland Cahokia/Grant Side Notched points (all chert unless otherwise noted). Haft Specimen* [D] C8-2-1 [D] D8-2-2 [D] C5-1-4 [D] D7-1-1 [D] C8-3-1 [D] F5-2-1 [D] C7-2-4 [D] C8-2-3 [D] D7-2-4 [D] C7-1-2 [D] D5-1-7 [BL] 1C-1(1)17 [BL] 3C-1(1)5 [BL] 40-1(1)2 [BL] 60-1(4)2 [BL] 3C-1(1)4 [BL] 1C-1(2)2 [BL] 2A-1(2)1 [BL] 1C-1(1)5 [BL] 70-1(3)1 [BL] 1C-1(1)16 [BL] WP5-13 [GG] C3/C4-20 Length Width (mm) (mm) 33.0 33.7 18.0 17.0 17.1 11.0 8.1 7.5 8.1 8.0 7.8 9.1 8.7 8.4 (g) (mm) 1.8 1.8 1.1 .6 .6 8.7 24.3 24.4 17.4 10.0 7.1 12.2 8.5 9.0 6.9 9.7 6.9 8.5 6.7 8.1 8.5 11.5 [GG] C3/C4-21 28.3 13.7 [GG] B6/C6-L-1 [GG] B6/C6-L-2 17.5 12.0 Average (N=26) Shoulder Weight Width Thickness 18.5 22.7 12.9 9.1 1.9 16.0 13.4 15.3 17.7 4.7 5.0 4.1 3.4 3.5 3.5 4.5 4.8 4.5 4.0 4.0 3.1 4.1 3.0 3.2 2.8 3.2 2.7 3.1 3.4 2.1 2.4 3.2 3.7 14.2 3.6 11.5 15.8 18.0 13.7 16.5 17.9 14.0 19.4 15.8 15.1 14.8 15.2 15.2 13.4 13.3 1.0 2.2 .9 .9 1.2 (mm) 14.3 14.1 14.0 15.2 3.3 3.6 Comments silicified sandstone silicified sandstone silicified sandstone; concave base silicified sandstone silicified sandstone * [D] Durst Rocksheiter [BL] Bard Lawrence Rocksheiter [GG] Gullickson's Glen through use. The Mill Pond contracting-stemmed points, like most of the 23 Waubesa points measured during this study, retain sinuous edges and show mini mal, if any, evidence of pressure flaking. Indeed, based on the common presence of multiple step fractures, these points appear to have been manufactured largely through percussion flaking (e.g., Salkin 1986:Figure 4.6), a practice Webb noted for some large, straight-stemmed points at Indian Knoll (Webb 1946:254). Not only do the Mill Pond bifaces give the impression of having been manufactured with expediency, but percussion flaking also can result in transverse "end shock" fractures in exactly the midblade position that occurs on the snapped specimens. This pattern is also apparent in the Florence Street and Bushmeyer site assem blages from Illinois (Emerson 1983:Figures 22,24; Emerson and Fortier 1986:Fig ures 18.12, 18.13; Morgan et al. 1986:Figure 11.10), Therefore, it is presumed This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms 52 Robert F. Boszhardt Table 10. Composite Measurements of Late Prehistoric Madison Triangular Points. Shoulder Width (mm) Thickness 1.0 16.5 15.2 16.2 17.7 14.5 15.5 14.6 14.6 16.5 17.5 4.5 3.7 4.1 5.4 3.5 3.4 3.5 3.2 4.0 4.0 silicified sandstone; 24.0 1.6 19.0 4.5 silicified sandstone; [GG] C-8-LL-3 15.0 .6 15.0 3.0 silicified sandstone; [GG] C-8-LL-1 15.0 .6 15.0 4.0 silicified sandstone; [GG] C-8-LL-2 [GG] C-8-LL-12 [GG] C-8-LL-11 [GG] C-8-LL-17 22.5 21.5 22.5 29.5 1.6 1.2 1.4 2.9 20.0 18.5 20.0 25.0 4.0 3.5 4.0 4.5 [GG] C-8-LL-16 [GG] D-LL-5 [GG] C-8-UL-1 [GG] B8-LL-3 [GG] B8-LL-3 20.5 21.0 15.5 29.3 30.5 1.1 1.1 .6 2.7 3.0 16.5 19.0 14.0 19.5 23.0 4.0 3.0 3.0 5.0 5.0 [GG] C3/C4-0-1 32.0 3.5 23.0 6.5 silicified sandstone; [GG] C3/C4-0-3 [GG] C3/C4-0-11 19.0 25.0 1.1 1.6 17.0 18.0 3.5 4.5 silicified sandstone [GG] C3/C4-0-4 [GG] C3/C4-0-12 [GG] C3/C4-0-2 18.5 20.0 25.0 .7 2.0 15.0 14.0 18.0 3.0 3.0 5.0 [GG] C3/C4-0-5 [GG] C3/C4-0-13 15.0 16.0 13.0 12.5 3.0 2.0 Length (mm) Weight [BL] WP5-73 [BL] 601(4)1 [BL] 6A-1(1)1 [BL] 4C-1(1)1 [BL] 10-1(1)1 [BL] 1E-1(2)1 [BL] WPS-74 [BL] 1C-1(2)1 [GG] I-L-4 [GG] I-L-l 28.5 23.6 26.2 28.2 21.4 23.5 22.6 22.0 21.0 18.0 1.8 1.2 1.5 1.9 1.0 1.2 1.0 0.9 [GG] I-L-3 Specimen* [GG] 07-UL-10 21.0 (8) 1.1 1.0 16.0 (mm) 3.5 Comments serrated serrated silicified sandstone silicified sandstone serrated serrated serrated serrated silicified sandstone Prairie du Chien chert Grand Meadow ? chert Prairie du Chien chert; very roughly made Prairie du Chien chert silicified sandstone silicified sandstone silicified sandstone silicified sandstone; serrated serrated Prairie du Chien chert; concave base silicified sandstone Prairie du Chien Chert silicified sandstone; concave base; very tip missing silicified sandstone Prairie du Chien chert; minimal flaking on ventral surface Prairie du Chien chert; missing small corner [GG] 07-UL-l 24.0 1.5 16.0 5.0 This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms of base silicified sandstone; concave base Contracting-Stemmed Points 53 Table 10. Continued. Shoulder Length Weight Width Thickness Specimen* (mm) (g) (mm) (mm) Comments [GG]07-UL-6 34.5 2.1 21.5 3.5 Prairie du Chien chert [GG]07-UL-7 18.0 1.0 16.0 3.5 Prairie du Chien chert; heat treated [GG]07-UL-9 25.5 1.4 17.5 3.5 Prairie du Chien chert [GG]07-UL-12 25.0 1.7 19.0 4.5 Galena? chert [GGJ07-UL-13 18.5 1.2 17.5 1.2 Prairie du Chien chert; heat treated; very tip missing [GG]07-UL-2 19.0 1.0 16.0 3.5 silicified sandstone [GG]07-UL-8 23.0 1.0 17.5 3.0 Prairie du Chien chert [GG]07-UL-11 23.5 1.7 17.5 3.5 Prairie du Chien chert; very tip missing [GG]07-UL-5 18. .8 15.5 3.0 silicified sandstone [GG] II-D-2 20. 1.2 19.0 3.0 silicified sandstone [GG] II-D-1 21.5 1.5 20.5 4.5 silicified sandstone [GG] II-D-3 25.0 1.8 20.5 4.5 silicified sandstone [GG] B-5-V-2 19.5 1.0 19.0 2.5 silicified sandstone; convex base [GG] II-D-3 26.5 1.8 19.5 3.5 Prairie du Chien chert; heat treated; convex base [GG] II-D-2 30.0 1.7 19.5 3.5 silicified sandstone; serrated [GG] II-D-1 18.0 .8 16.5 3.5 silicified sandstone; very small portion of base missing at corner [GG]07-LL-2 16.5 .9 17.5 3.5 Prairie du Chien chert [GG] B3-B4-L-13 24.5 2.2 21.2 5.0 silicified sandstone [GG] B3-B4-L-15 31.0 2.2 20.5 5.0 silicified sandstone; serrated [GG] B3-B4-L-14 25.0 2.2 20.0 4.5 silicified sandstone [GG] B3-B4-L-8 18.0 .9 16.0 3.5 silicified sandstone [GG]C-5-UL-l 30.0 2.2 21.0 3.5 silicified sandstone; serrated [GG]C-5-UL-5 16.0 .8 14.0 4.0 silicified sandstone; concave base [GG]C-8-U-l 35.5 3.6 25.0 4.0 Prairie du Chien chert; refit and glued frags. [GG]B5-L-2 26.0 1.7 18.0 4.5 silicified sandstone [GG]B5-L-1 17.0 .5 12.0 3.5 silicified sandstone; very small (reworked?) [GG]B5-L-4 26.5 1.7 16.0 4.0 silicified sandstone [GG]B5-L-5 19.0 .9 16.5 3.0 silicified sandstone [GG] II-D-1 25.0 1.3 15.5 3.5 silicified sandstone; serrated [GG]II-D-5 23.5 1.6 18 4.0 silicified sandstone [GG]II-D-4 19.0 1.1 17.5 3.5 silicified sandstone This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms 54 Robert F Boszhardt Table 10. Continued. Shoulder Specimen* Length Weight Width Thickness (mm) (g) (mm) (mm) Comments [GG]II-D-3 19.5 1.7 19.0 4.0 silicified sandstone; concave base [GG]II-D-2 25.0 1.8 21.0 5.5 silicified sandstone; concave sides [GG]C3/C4-2 38.5 7.3 30.2 7.0 silicified sandstone; very large [GG]D-8-L-l 18.0 1.1 17.5 3.7 silicified sandstone; serrated [GGJD-8-L-7 19.7 .8 15.9 3.3 Grand Meadow chert [GG]D-8-L-2 16.2 .7 14.9 3.5 silicified sandstone [GG] A8/B8-D-1 22.6 1.5 21.1 4.2 silicified sandstone; flake point w/ minimal retouch on ventral surface [GG]II-D-1 30.2 2.4 21.2 4.0 silicified sandstone; concave base [GG]II-D-4 20.4 1.3 16.4 3.9 silicified sandstone; deep concave base [GG]II-D-5 22.5 1.3 19.6 3.2 silicified sandstone [GG]II-D-2 19.5 1.1 17.0 4.4 silicified sandstone [GG]II-D-3 20.0 .9 14.0 3.6 silicified sandstone [GG]II-D-9 20.9 1.1 18.5 4.2 Prairie du Chien chert; slightly eared [GG]C7-UL-6 20 1.1 14 4.2 Prairie du Chien chert; heat treated [GG]C7-UL-7 30(+) 2.4 21.9 3.7 Prairie du Chien chert; very tip is missing [GG]C7-UL-1 21.6 1.1 18.5 3.5 silicified sandstone [GG]B7-U-1 26.8 1.8 17.8 4.0 silicified sandstone; concave base [GG]IV-U-2 22.5 1.4 18.5 3.9 silicified sandstone [GG]D6-UL-16 20.0 .8 13.2 2.7 silicified sandstone [GG]D6-UL-9 26.4 1.0 12.4 3.3 Grand Meadow chert [GG]D6-UL- 25.4 2.3 16.9 5.4 silicified sandstone [GG]II-D-1 24.2 1.3 14.5 3.7 Prairie du Chien chert [GG]C3/C4-9 24.5 2 20.6 4.3 silicified sandstone [GG]C3/C4-23 27.9 .7 13.1 3.3 Prairie du Chien chert; convex base [GG]C3/C4-3 24.1 1.8 19.1 4.0 silicified sandstone [GG]C3/C4-8 23.5 1.4 16.1 4.5 silicified sandstone; serrated [GG]C3/C4-24 16.8 .5 10.8 2.9 Cochrane chert [GG]C3/C4-4 22.8 1.6 17.7 4.2 silicified sandstone [GG]C3/C4-7 24.7 1.2 13.7 3.7 silicified sandstone [GG]C3/C4-5 17.2 .6 13.1 3.3 silicified sandstone; serrated [GG]C7-LL-2 28.1 1.6 16.5 3.8 silicified sandstone; serrated This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms Contracting-Stemmed Points 55 Table 10. Continued. Shoulder Specimen* Length Weight Width Thickness (mm) (g) (mm) (mm) Comments [GG] C7-LL-1 [GG] C7-LL-3 [GG] C7-LL-8 24.8 21.9 16.2 1.1 1.2 15.7 3.2 silicified sandstone 16.3 3.8 silicified sandstone [GG] C7-LL-7 [GG] B8-UL-1 [GG] B8-UL-2 18.1 .9 1.6 .6 1.6 1.9 16.7 3.3 Prairie du 17.8 3.6 silicified 14.6 3.4 silicified 16.8 3.9 silicified 20.5 3.4 silicified 25.5 15.1 .7 12.7 3.5 unknown chert; concave base Chien chert sandstone sandstone sandstone sandstone [GG] B6/C6-L-4 26.0 29.6 16.3 [GG] B6/C6-L-3 [GG] B6/C6-L-4 22.1 19.5 1.3 17.2 4.0 silicified sandstone [GG] B6/C6-L-5 [GG] B6/C6-L-11 [GG] B6/C6-L-10 26.2 22.3 21.9 22.8 1.5 1.3 16.7 1.6 20.5 3.1 silicified sandstone 18.9 4.0 silicified sandstone 1.2 4.6 silicified sandstone 17.2 3.8 [GG] B7-disturbed [GG] B6/C6-L-6 Averages (N=108) 15.6 3.9 silicified sandstone; concave base 13.6 3.8 silicified sandstone; concave base & sides *[BL] Bard Lawrence Rocksheiter (whole points only) [GG] Gullickson's Glen Rocksheiter (whole or nearly complete point Table 11. Summary of Averages for Points in Chronolog Length Haft Weight Shoulder Thickness Type (mm) Width (mm) (g) Width (mm) (mm) Price (N=12) 52.06 17.43 10.89 24.03 9.61 Raddatz (N=40) 46.39 19.51 9.44 26.48 8.95 Preston (N=33) 33.44 11.84 4.22 21.88 7.34 Durst (N=100) 37.50 10.97 4.65 18.64 7.40 Kramer (N=16) 54.10 16.83 10.20 22.01 9.27 Waubesa (N=30) 57.85 17.79 13.21 26.29 9.85 Honey Creek (N=20) 20.86 11.12 1.31 14.95 3.98 Cahokia (N=26) 22.65 9.14 1.24 15.16 3.57 Madison (N=108) 22.78 - 1.56 17.22 3.82 that the Mill Pond bifaces were broken in manufacture rather than during use knives. Blodgett (2001) undertook controlled experiments to evaluate whether con tracting-stemmed bifaces may have functioned as knives. He hafted a series of replicated contracting-stemmed bifaces to notched wooden handles using adh? sives of varying tensile strength (360 to 3,000 PSI); one was lashed only with deer sinew. Each replicated knife haft was wrapped in sinew and then coated This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms 56 Robert F Boszhardt tr o r- to ?2 ; ? & Cl O <?_ <: _ ~ ? _ ? _ u -g E g E ?S % E SE <S O g CO ^ ?JD E This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms Contracting-Stemmed Points 57 Figure 3. Waubesa contracting-stemmed points from the Mill Pond site. Note midblade transverse fractures in lower row, suggesting end-shock during percus sion flaking. with natural pine pitch to keep the sinew from becoming wet. All knives were used to cut vegetables, hide, meat, and wood for 100 strokes. All of the knives hafted with adhesive held but were found to be inefficient cutting tools. Because of their sinuous edges and thick blades, these knives tended to rip and tear mate rials. In contrast, an unmodified flake cut much more cleanly in every case. The knife hafted only with deer sinew weakened and slipped from the haft while cutting meat. The only damage to the replicated bifaces was the fracturing of one tip while cutting a potato. Blodgett (2001) also conducted a supplementary test that compared the maxi mum lateral pressure of two replica contracting-stemmed knives, one hafted to a socketed wood handle with a commercial glue rated at a tensile strength of 360 PSI and one mounted with natural pine pitch. The knife hafted with commercial glue failed when lateral pressure of 47.8 kg (105.5 lb) was applied. This failure happened when the stem broke 5 mm below the shoulder. The haft of the knife mounted with natural pine pitch failed at 71.5 kg (157.5 lbs) of lateral pressure, but the failure did not involve breakage of the biface. These experiments reveal that contracting-stemmed bifaces could function as knives, particularly if hafted with a mastic such as pine pitch. However, the sinu ous edge, typical of many contracting-stemmed artifacts, is far less efficient than fresh flakes for cutting. Furthermore, the biface tip and stem fractures that oc curred during the experiments did not duplicate the midblade fractures noted for Waubesa points at the Mill Pond site and for several contracting-stemmed as This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms 58 Robert F. Boszhardt semblages in Illinois. Compound Shafts Why then did point styles evolve from large side-notched dart forms to small expanding-stemmed forms to large contracting-stemmed forms during the Ar chaic-Woodland transition? The hypothesis offered here is that Middle Archaic side-notched tips were fixed directly to main spear shafts, a configuration that gave way in Late Archaic times to compound darts with wood foreshafts armed with "small" points that resemble arrow tips in size. These weapons, in turn, were replaced at the onset of the Woodland tradition by larger contracting stemmed point tips that, in effect, eliminated the need to manufacture (and inevi tably lose some) wooden foreshafts. Preserved dart foreshafts have been recovered from several dry caves in the Ozarks region (e.g., Harrington 1960; Shippee 1966). Harrington (1960:Plate XXV) illustrates three complete wooden foreshafts from the Bushwhack, Elk Spring, and White Bluff Rockshelters. The lengths range from 7.75 to 8.625 inches (approximately 18.5-21.5 cm), and each is about 1 cm in diameter. Shippee (1966:Figure 9) illustrates two foreshafts from Arnold Research Cave in Mis souri that are approximately 16 and 19 cm long and also just under a centimeter in diameter. The length range of the six preserved Ozark rocksheiter foreshafts reported by Harrington and Shippee is 16.0-21.5 cm, for an average of 18.25 cm. Their di ameters are consistently about 1 cm. Weights are not reported, but a fresh 18.25 cm-long 1-cm-diameter commercial dowel weighs 8.2 g. Assuming that Preston and Durst points represented small dart tips attached to such foreshafts, combin ing the average weights of these points (4.22 and 4.65 g) with the assumed 8.2 g weight of an average foreshaft results in combined tip weights of 12.42 g and 12.85 g respectively. If applied to the point weight pattern depicted in Figure 4, the Late Archaic "valley" is filled, creating a continuous weight plateau at or above 10 g from the Early Archaic through Early Woodland. The first significant weight change, then, coincides with the advent of small late prehistoric point tips that weigh consistently less than 2 g. Harrington (1960) and Shippee (1966) illustrate points that were still attached to their foreshafts when recovered. Harrington (1960:Plate XXVc) shows an expanding stem form with a sharp shoulder from the Bushwhack Rocksheiter. Its length is approximately 35 mm and the haft width is about 15 mm. The point hafted onto a foreshaft from Arnold Research Cave is a weakly side-notched variety, approximately 40 mm long and with a haft width of about 18 mm. Harrington (1960:Plate XXIV) also illustrates four unhafted "dart points" from the Bluff Dweller sites. One is a straight- or slightly expanding-stemmed form with sharp shoulder barbs, whose length is 1.875 inches (approximately 47 mm). A second point is corner notched and 1.5 inches (38 mm) long. The final two unhafted points have contracting stems, and both are larger, each measuring 2.5 inches (65 mm) in length. Thus, in this small set, notched points and those hafted This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms Contracting-Stemmed Points 59 c o 1 ....,.?.?.?^..,........_._........___ ?fl I ! : : 5 CO :1^ 1 i ?11 I _____ a> \ _. ^ = q -=_= = ?~? . 3 s ce _____ ^ s ^^^^eI^ ------ r}- CN CO CO " CM O ^ ? SLUEJ6 .SP This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms fa 60 Robert F Boszhardt onto foreshafts are substantially smaller than contracting-stemmed points. In a more extensive analysis, Thomas (1978) measured 10 archaeological atlatl darts that retain hafted points, all from southwestern sites. The dart tips ranged in length from 34.0 to 65.4 mm and in weight from 2.0 to 7.9 g. Point form was not presented. The wooden foreshafts ranged in length from 56.9 to 171.5 mm, and diameters ranged from 8.5 to 12.3 mm. Notably, Thomas's analysis revealed that "larger foreshafts tend to have smaller points" (1978:469). Although Thomas characterized this result as "unexpected," it makes perfect sense when consider ing the combined weight of the tip and foreshaft. As Pyszczyk (1999) estab lished in his review of ethnographic bows and arrows, alteration of any compo nent requires a compensatory adjustment to others for balance and flight. In the case of darts, it is to be expected that point size would diminish as foreshaft size increased and vice versa. Thomas's observation further supports the hypothesis offered here, that is, reduced point size in the Late Archaic may reflect compen sation for the added weight of wooden foreshafts. More specifically, weight ad justments of the dart tip shifted from large (Raddatz) chipped stone points to smaller Preston and Durst tips that were likely mounted in wooden foreshafts. These forms subsequently were replaced by larger Kramer and Waubesa points that served the function of composite foreshafts/point tips, but saved time and labor. Shott (1997) expanded on Thomas's ethnoarchaeological study by analyzing 30 museum-curated dart specimens, most from the American Southwest. His results largely corroborated those of Thomas, and both researchers developed multivariate classification functions to assist in evaluating whether archaeologi cal points were used as dart or arrow tips. The mean data for the Driftless Area points presented in Table 11 were subjected to both Thomas's and Shott's classi fication functions with the following results. Thomas's formula placed Price stemmed, Raddatz side-notched, Preston corner-notched, and Waubesa contract ing-stemmed within the dart tip category, while Durst and Kramer points, along with Honey Creek corner-notched and Cahokia points, were classified as arrow tips. Applying Shott's four-variable formula to the data set revealed that Price, Raddatz, and Waubesa correlate with dart tips, while Preston and Durst again join Honey Creek and Cahokia points within the arrow tip category. Shott's func tion had Kramer points producing an equal value for both dart and arrow formu las. Neither Thomas nor Shott included the weight attribute within their classifi cation function formulas, but, as noted earlier, weight differences are substantial between the Late Archaic Preston and Durst types and the late prehistoric arrow tip types (Honey Creek, Cahokia, and Madison). Unnotched Madison points do not lend themselves to Thomas's or Shott's analysis because haft width (neck) is absent. Shott concluded that shoulder width alone is as effective as any multivariate function in discriminating dart points from arrow tips. Specifically, he found the mean shoulder width of dart tips to be approximately 23 mm, in contrast to ar row tips, which exhibited a mean shoulder width of about 14.5 mm (Shott 1997:91). In the Driftless Area data set presented here, shoulder width is the This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms Contracfmg-Stemmed Points 61 least variable attribute among the Archaic and Early Woodland types, including the otherwise smaller Preston and Durst points (average range 21.88-18.64 mm). This finding supports the interpretation that these types also served as dart tips. Shoulder width on the three late prehistoric arrow tip types from the Driftless Area is smaller, with averages ranging from 14.15 to 17.22 mm, the latter repre senting the base of unnotched triangular forms. Conclusion The relatively small projectile points manufactured during the Late Archaic in parts of eastern North America might represent early examples of bow and ar row technology. However, in the Driftless Area, at least, these points are sub stantially larger and heavier than late prehistoric arrow tips. Because the Late Archaic types were replaced by larger straight- and then contracting-stemmed dart tips, the "small" Archaic forms may well represent tips mounted to dart foreshafts. This suggestion is supported by a few examples of dart foreshafts hafted with relatively small tips recovered from dry caves in the Ozarks region and by Thomas's observation of inverse relations between foreshaft and tip sizes. When the assumed weight of average-sized foreshafts is added to average weights of small Late Archaic point types such as Preston and Durst, the combined weight is comparable to preceding and succeeding larger dart types. In contrast, virtu ally all late prehistoric points weigh less than 2 g. This pattern strongly supports the conclusion that adoption and retention of bow and arrow technology oc curred in late prehistoric times, beginning with the Honey Creek type. The vast majority of contracting-stemmed projectile points in southern Wis consin are relatively thick, have multiple step fractures (suggesting manufacture through hard hammer percussion), and retain sinuous edges from not having been pressure flaked. These points appear to have been manufactured with expe diency, and they were therefore likely expendable. The contracting-stem design would facilitate removal, allowing the point to remain in the wound and the shaft to be recovered and easily retipped. The archaeological implications of such an interpretation are that straight- and contracting-stemmed points should be widely scattered on the landscape from loss during hunting and common only at camp locations where they were being manufactured. Farnsworth and Asch (1986:407) documented several hundred Belknap points as isolated finds throughout the Lower Illinois Valley uplands, suggesting loss through hunting. In contrast, the contracting-stemmed points found in concentrations at Beach, Mill Pond, and numerous other floodplain camps along the Upper Mississippi River are typi cally made of local cherts, and many are fractured as if broken in manufacture. The near absence of contracting-stemmed points from Driftless Area rockshelters may not necessarily indicate an absence of Early Woodland occupation but, rather, that such points were neither manufactured at nor returned to those winter base camps. In contrast, the common presence of Raddatz, Preston, and Durst points in rockshelters may attest to their being returned to the camps, their notched bases having been lashed to shafts. Indeed, many of the Raddatz points from the This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms 62 Robert F Boszhardt Raddatz Rocksheiter are broken or heavily resharpened (Wittry 1959a) and per haps were discarded only after the binding was cut or untied to remove them from their reusable hafts. Experiments found that contracting-stemmed bifaces can function as hafted knives but are effective only if the haft is set with an adhesive such as pine pitch. Furthermore, the thick blade and sinuous edges found on most contracting stemmed bifaces (e.g., Waubesa points) produces a rough cut that is far less clean than slices easily produced on the same materials by unworked flakes. Nonetheless, some contracting-stemmed bifaces in the archaeological record (e.g., Goose Lake and Peisker knives, as well as points classified as "Dickson" in southern Wisconsin) have edges that were straightened through pressure flake retouching, and these appear to have served as hafted knives. Avenues for future research on this topic include additional experiments to evaluate flight and penetration characteristics of small, notched tips fixed to foreshafts and of larger, straight- and contracting-stemmed points mounted di rectly to dart shafts. It is also intriguing that the demise of bannerstones, often thought to have served as atlatl weights, coincided almost exactly with the tran sition from Archaic to Woodland in the Eastern Woodlands (Knoblock 1939; Kwas 1981). Consequently, experiments should consider the effects of using bannerstones with various point-and-foreshaft combinations, including variations such as rigid versus flexible shafts. Other potential lines of research may follow the path suggested by Nassaney and Pyle (1999:260) in looking at social, political, and economic factors influ encing the adoption of new technologies. For example, the 500 years preceding European contact were dominated by the Madison triangular form. Like Waubesa contracting-stemmed points, triangular arrow tips are relatively easy to manu facture and to replace. Similarly, triangular forms were apparently intended to stay in the wound, allowing recovery of the much more difficult to manufacture arrow shaft. Like the contracting-stemmed points associated with skeletal re mains at Indian Knoll, the detachable Madison point offered an obvious advan tage in warfare. Indeed, several ethnographic sources segregate war points from hunting points on the basis of hafts designed to allow the point to remain in the wound of an enemy (Pyszczyk 1999:169). Finally, any attempt to correlate evolutionary patterns of point form and tech nological change inevitably raises questions of temporal control and stylistic homogeneity. Historical accounts indicate that some Native American groups used multiple arrow tip styles for various functions, e.g., for hunting different types of game or for warfare (Pyszczyk 1999). Individual prehistoric cultures also might have manufactured a variety of projectile point "types" at any one time. Indeed, many point types are attributed to lengthy periods of manufacture by multiple cultures, their ages often overlapping the age estimates of other types. Waubesa points, for example, were originally attributed to Late Archaic through Middle Woodland cultures in southern Wisconsin (Goldstein and Osborn 1988:44), and Nassaney and Pyle (1999:247) suggest a similar lengthy range for Gary points at Plum Bayou. However, most archaeological interpretations of This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms Contracting-Stemmed Points 63 this sort are based on contexts that, at the very minimum, allow the possibility of temporal mixing. These contexts commonly include surface associations (e.g., Nassaney and Pyle 1999) and shallow, unstratified deposits or mound fill (e.g., Hurley 1975). Even stratified rocksheiter deposits frequently exhibit mixing from natural (animal burrowing) or cultural activities (pit excavation) (e.g., Ahler 1993). Short-term cultural deposits in environments that are rapidly sealed and not subject to subsequent disturbances allow a better evaluation of the range of spe cific tool forms at any one time. The Mill Pond site represents a short-term Prai rie phase occupation that was sealed by subsequent Mississippi River flood de posits. The only point type associated with the Prairie phase shell midden in block excavations at Mill Pond was Waubesa contracting-stemmed, represented by 10 specimens (Theler 1987:76). The 41 zones within 18 ft of deposits at the deeply stratified St. Albans site in West Virginia reflect alternating cultural zones separated by sterile flood deposits (Broyles 1971:14-20). At St. Albans, a vari ety of point types were recovered spanning thousands of years within the Ar chaic tradition, yet "each type of projectile point was confined to one or two zones which were separated from the zones above and below by bands of sterile clay or sand" (Broyles 1971:24). Therefore, in some midcontinental archaeo logical contexts with unquestionable stratigraphie integrity, specific projectile point types have been found to be isolated in time. The evolutionary technologi cal hypothesis offered here is based on a point sequence that is supported by radiocarbon and stratigraphie evidence in the Driftless Area but remains to be validated by future research. Acknowledgments My appreciation to Nancy Hoffman and Denise Wiggins of the Museum Divi sion of the Wisconsin Historical Society for making available the point assem blages from the Durst, Gullickson's Glen, Lawrence, and Raddatz Rockshelters for measurement and photography. James Stoltman of the University of Wiscon sin-Madison likewise provided access to the collections from the Beach and Mill Pond sites as well as the Preston Rocksheiter. Several University of Wis consin-La Crosse undergraduate archaeology majors assisted in the research, including Dillon Carr, who measured the Gullickson's Glen points, Dustin Blodgett and Travis Traverna, who conducted experimental studies on contract ing-stemmed bifaces, and Beth Wergin, who compiled references during the ini tial stages of the research. The manuscript has been steadily enhanced through comments provided by a number of reviewers, including Michael Shott, Mark Lynott, and William Green, along with two anonymous readers. Finally, thanks to Brad Perkl for inviting my participation in the Early Woodland symposium at the 2000 Midwest/Plains Conference in St. Paul. Mississipi Valley Archaeology Center 1725 State St. La Crosse, WI54601 This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms 64 Robert F Boszhardt References Ahler, Steven R. 1993 Stratigraphy and Radiocarbon Chronology of Modoc Rocksheiter, Illinois. American Antiquity 58:462-487. Amick, Daniel S. 1994 Technological Organization and the Structure of Inference in Lithic Analysis: An Examination of Folsom Hunting Behavior in the American Southwest. In The Organization of North American Prehistoric Chipped Stone Tool Tech nologies, edited by Philip J. Carr, pp. 9-34. Archaeological Series 7. Interna tional Monographs in Prehistory, Ann Arbor, Michigan. Amick, Daniel S., Robert F. Boszhardt, Ken Hensel, Matthew G. Hill, Thomas J. Loebel, and Dean W. Wilder 1999 Pure Paleo in Western Wisconsin. Reports of Investigations No. 350. Missis sippi Valley Archaeology Center, University of Wisconsin-La Crosse. Benn, David W. 1979 Some Trends and Traditions in Woodland Cultures of the Quad State Region of the Upper Mississippi River Basin. The Wisconsin Archeologist 60:47-82. Blitz, John H. 1988 Adoption of the Bow in Prehistoric North America. North American Archae ologist 9:123-145. Blodgett, Dustin J. 2001 Exploring the Possible Use of Contracting Stem Projectile Points as Hafted Knife Blades. Senior Thesis for a major in Archaeology. University of Wis consin-La Crosse. Bradbury, Andrew P. 1997 The Bow and Arrow in the Eastern Woodlands: Evidence for an Archaic Ori gin. North American Archaeologist 18:207-233. Broyles, Bettye J. 1971 Second Preliminary Report: The St. Albans Site, Kanawha County, West Vir ginia. Report of Archaeological Investigations No. 3. West Virginia Geologi cal and Economic Survey, Morgantown. Conrad, Lawrence A. 1986 The Late Archaic/Early Woodland Transition in the Interior of West-Central Illinois. In Early Woodland Archeology, edited by Kenneth B. Farns worth and Thomas E. Emerson, pp. 301-325. Center for American Archeology Press, Kamps ville, Illinois. Conrad, Lawrence A., Susan L. Gardner, J. Joe Alford 1986 A Note on the Late Archaic, Early Woodland, and Early Middle Woodland Occupations of the Lima Lake Locality, Adams and Hancock Counties, Illi nois. In Early Woodland Archeology, edited by Kenneth B. Famsworth and Thomas E. Emerson, pp. 191-206. Center for American Archeology Press, Kamps ville, Illinois. Corliss, David W. 1980 Arrowpoint or Dart Point: An Uninteresting Answer to a Tiresome Question. American Antiquity 45:351-352. This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms Contracting-Stemmed Points 65 Emerson, Thomas E. 1983 The Florence Street Site. American Bottom Archaeology FAI-270 Site Reports Vol. 2. University of Illinois Press, Urbana. 1986 A Retrospective Look at the Earliest Woodland Cultures in the American Heart land. In Early Woodland Archeology, edited by Kenneth B. Farnsworth and Thomas E. Emerson, pp. 621-633. Center for American Archeology Press, Kamps ville, Illinois. Emerson, Thomas E., and Andrew C. Fortier 1986 Early Woodland Cultural Variation, Subsistence, and Settlement in the Ameri can Bottom. In Early Woodland Archeology, edited by Kenneth B. Farnsworth and Thomas E. Emerson, pp. 475-519. Center for American Archeology Press, Kamps ville, Illinois. Farnsworth, Kenneth ., and David L. Asch 1986 Early Woodland Chronology, Artifact Styles, and Settlement Distribution in the Lower Illinois Valley Region. In Early Woodland Archeology, edited by Kenneth B. Farnsworth and Thomas E. Emerson, pp. 326-457. Center for American Archeology Press, Kampsville, Illinois. Freeman, Joan E. 1966 Price Site III, Ri 4, a Burial Ground in Richland County, Wisconsin. The Wis consin Archeologist 47:33-75. Goldstein, Lynne G., and Sannie K. Osborn 1988 A Guide to Common Prehistoric Projectile Points in Wisconsin. Milwaukee Public Museum, Milwaukee, Wisconsin. Halsey, John R. 1971 Report on the 1970 Excavations in the Proposed La Farge Reservoir. Report submitted to the National Park Service. On file, State Historical Society of Wisconsin, Madison. 1976 Report of the Investigations in the LaFarge Lake Project Area: 1973 and 1974 Seasons. Report on file, State Historical Society of Wisconsin, Madison. Harrington, M. R. 1960 The Ozark Bluff Dwellers. Museum of the American Indian, Heye Founda tion, New York. Huelsbeck, David R. 1974 Archeological Survey in the Proposed Upland Recreation Areas of the Lake LaFarge Project. Report on file, State Historical Society of Wisconsin, Madi son. Hurley, William M. 1965 Archaeological Research in the Projected Kickapoo Reservoir, Vernon County, Wisconsin. The Wisconsin Archeologist 46:1-114. 1975 An Analysis of Effigy Mound Complexes in Wisconsin. Anthropological Pa pers No. 59. Museum of Anthropology, University of Michigan, Ann Arbor. Justice, Noel D. 1987 Stone Age Spear and Arrow Points of the Midcontinental and Eastern United States: A Modern Survey and Reference. Indiana University Press, Bloomington. Knoblock, Byron W 1939 Bannerstones of the North American Indians. Privately published, La Grange, Illinois. This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms 66 Robert R Boszhardt Kwas, Mary L. 1981 Bannerstones as Chronological Markers in the Southeastern United States. Tennessee Anthropologist 6:144-171. Martin, Lawrence 1965 The Physical Geography of Wisconsin. University of Wisconsin Press, Madi son. Mead, Barbara 1975 Report on the Plum Run Archeological Survey of the La Farge Lake Project. Report on file, State Historical Society of Wisconsin, Madison. 1976a Archeological Investigations in Recreation Areas C and H of the La Farge Lake Project. Report on file, State Historical Society of Wisconsin, Madison. 1976b Archeological Investigations in the Lower Pool Area of the La Farge Lake Project Area. Report on file, State Historical Society of Wisconsin, Madison. 1979 The Rehbein I Site (47-R?-81): A Multicomponent Site in Southwestern Wis consin. The Wisconsin Archeologist 60:91-182. Morgan, David T., David L. Asch, and C. Russell Stafford 1986 Marion and Black Sand Occupations in the Sny Bottom of the Mississippi River. In Early Woodland Archeology, edited by Kenneth B. Farns worth and Thomas E. Emerson, pp. 207-230. Center for American Archeology Press, Kamps ville, Illinois. Nassaney, Michael S., and Kendra Pyle 1999 The Adoption of the Bow and Arrow in Eastern North America: A View from Central Arkansas. American Antiquity 64:243-263. Odell, George H. 1988 Addressing Prehistoric Hunting Practices through Stone Tool Analysis. Ameri can Anthropologist 90:335-356. Price, James E. 1986 Tchula Period Occupation along the Ozark Border in Southeast Missouri. In Early Woodland Archeology, edited by Kenneth B. Famsworth and Thomas E. Emerson, pp. 535-545. Center for American Archeology Press, Kampsville, Illinois. Pyszczyk, Heinz W. 1999 Historic Period Metal Projectile Points and Arrows, Alberta, Canada: A Theory for Aboriginal Arrow Design on the Great Plains. Plains Anthropologist 44:163 187. Salkin, Philip H. 1986 The Lake Farms Phase: The Early Woodland Stage in South-Central Wiscon sin as Seen from the Lake Farms Archeological District. In Early Woodland Archeology, edited by Kenneth B. Famsworth and Thomas E. Emerson, pp. 92-120. Center for American Archeology Press, Kampsville, Illinois. Seeman, Mark F. 1992 The Bow and Arrow, the Intrusive Mound Complex and a Late Woodland Jack's Reef Horizon in the Middle Ohio Valley. In Cultural Variability in Con text: Woodland Settlements of the Mid-Ohio Valley, edited by Mark Seeman, pp. 41-51. Kent State University Press, Kent, Ohio. Shippee, J. Mett 1966 Arnold Research Cave. Special Publication 1. Missouri Archaeologi This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms Contracting-Stemmed Points 67 cal Society, Columbia. Shott, Michael J. 1993 Spears, Darts, and Arrows: Late Woodland Hunting Techniques in the Upper Ohio Valley. American Antiquity 58:425-443. 1997 Stones and Shafts Redux: The Metric Discrimination of Chipped-Stone Dart and Arrow Points. American Antiquity 62:86-101. Stevenson, Katherine P., Robert F. Boszhardt, Charles R. Moffat, Philip H. Salkin, Thomas C. Pleger, James L. Theler, and Constance A. Arzigian 1998 The Woodland Tradition. In Wisconsin Archaeology, edited by Robert A. Bir mingham, Carol I. Mason, and James B. Stoltman, pp. 140-201. The Wiscon sin Archeologist 78 (for 1997). Stoltman, James B. 1990 The Woodland Tradition in the Prairie du Chien Locality. In The Woodland Tradition in the Western Great Lakes: Papers Presented to Eiden Johnson, edited by Guy E. Gibbon, pp. 239-259. Publications in Anthropology No. 4. University of Minnesota, Minneapolis. 1998 The Archaic Tradition. In Wisconsin Archaeology, edited by Robert A. Bir mingham, Carol I. Mason, and James B. Stoltman, pp. 112-139. The Wiscon sin Archeologist 78 (for 1997). Storck, Peter 1967 Upland Surface Archaeological Project, Kickapoo River Valley, Vernon County, Wisconsin. Unpublished M.A. thesis, Department of Anthropology, Univer sity of Wisconsin-Madison. Theler, James L. 1986 The Early Woodland Component at the Mill Pond Site, Wisconsin. In Early Woodland Archeology, edited by Kenneth B. Farnsworth and Thomas E. Emerson, pp. 137-158. Center for American Archeology Press, Kampsville, Illinois. 1987 Woodland Tradition Economic Strategies: Animal Resource Utilization in Southwestern Wisconsin and Northeastern Iowa. Report No. 17. Office of the State Archaeologist, University of Iowa, Iowa City. Thomas, David Hurst 1978 Arrowheads and Atlatl Darts: How the Stones Got the Shaft. American Antiq uity 43:461-472. Webb, WilliamS. 1946 Indian Knoll, Site Oh 2, Ohio County, Kentucky. Reports in Anthropology Vol. 4, No. 3, Part 1. University of Kentucky, Lexington. 1950 The Carlson Annis Mound Site 5, Butler County, Kentucky. Reports in Anthro pology Vol. 7, No. 4. University of Kentucky, Lexington. Winters, Howard D. 1969 The Riverton Culture. Report of Investigation No. 13. Illinois State Museum, Springfield. Wittry, Warren L. 1959a The Raddatz Rocksheiter, Sk5, Wisconsin. The Wisconsin Archeologist 40:33 69. 1959b Archeological Studies of Four Wisconsin Rockshelters. The Wisconsin Arche ologist 40:137-267. This content downloaded from 128.104.46.196 on Sat, 30 Jan 2021 01:30:34 UTC All use subject to https://about.jstor.org/terms