The millennia-long use history of triangular bifaces

Abstract

Relatively small, triangular bifaces often considered to be projectile points have a demonstrable use history that includes the Middle Archaic, Late Archaic, Early Woodland, late Middle Woodland, Late Woodland, and Contact periods of regional archaeology. Radiocarbon dates and other data are used to document this extensive history using the Upper Delaware Valley of New Jersey, Pennsylvania, and New York as a case study. Observed trends are evaluated in a broader regional context. The degree to which triangles of different ages can be distinguished from one another is addressed and suggestions for future research are made.

Keywords

Triangular bifaces and radiocarbon dates Middle Atlantic Archaeology Delaware Valley Native American archaeology

Introduction and background

The fact that relatively small triangular bifaces, often presumed to be projectile points, were in use prior to the Late Woodland period of the Middle Atlantic and Northeastern regions was established long ago (e.g., Funk et al., 1971; Leslie, 1963; Ritchie, 1961, 1971; Stewart and Cavallo, 1991). Yet these bifaces continue to be relegated to Late Woodland times when assigning age to otherwise undatable deposits in site-specific studies, and in analyses that employ surface and plowzone sites in defining settlement patterns. Sites in the Delaware Valley, Middle Atlantic, and Northeast regions provide evidence that triangular points are not limited to assemblages post-dating 600/900 AD. They also are known to be part of Middle Archaic, Late Archaic, and Early Woodland assemblages. Without a clear understanding of these earlier triangles, we run the risk of misusing typology in assigning age to deposits, especially surface and plowzone sites employed in the analysis and interpretation of the Native American past.

Here I use radiocarbon dates and other data to document the extensive history of triangular bifaces using the Upper Delaware Valley of New Jersey, Pennsylvania, and New York as a case study. The history of relevant research is summarized and morphological and functional data are included when available. Discussions are framed within the context of the data base for the broader region, and the history of research dealing with triangular bifaces. The degree to which triangles of different ages can be distinguished from one another is addressed as are directions that future research might take. The case study derives from a larger alternative mitigation project sponsored by the New Jersey Historic Preservation Office (Stewart, 2018).

For the purpose of this endeavor, the Upper Delaware Valley is defined by portions of the drainage basin that exist in the following states and counties (Figure 1): Warren and Sussex counties, New Jersey; Orange, Sullivan, Delaware, and Broome counties, New York; and Northampton, Monroe, Pike, and Wayne counties, Pennsylvania. The defined area is much larger than what has often been considered as the Upper Delaware by archaeologists in the past (cf. Custer, 1996; Kinsey, 1972; Kraft, 2001). The Pennsylvania State Historic Preservation Office includes Northampton County in what it considers to be the Upper Delaware Valley, which extends the geographic boundary well south of the Delaware Water Gap. I have included Northampton County here to complement the downriver extent of Warren County in New Jersey. The degree to which all, or portions of this larger area corresponds with Pre-contact cultural or group territories remains an open question contingent on a variety of diachronic analyses. Evaluating any data set from a broader geographic context is necessary to examine existing assumptions about the correspondence of natural and cultural areas, and to better understand patterns of social interaction and cultural change. Environmental scientists have used the designation, Upper Delaware Valley, in very different ways (e.g., Delaware River Basin Commission, 2018; Goetz et al., 2011: 18; Pennsylvania Fish and Boat Commission, 2011:37; USGS National Water Information System, 2018).

Figure 1.

Upper Delaware Valley project area showing relevant counties of New Jersey, Pennsylvania, and New York. Not labeled is Otsego County situated to the north of Delaware County, New York. Areas within the Delaware River Basin are shown in white. Modified from a map by the Delaware River Basin Commission (2018).

The radiocarbon data base

Data marshaled for this report were derived from a review of all cultural resource management (CRM) reports for Warren and Sussex counties on file at the New Jersey Historic Preservation Office as of 2016. Select Phase I and all Phase II and Phase III CRM reports for New York and Pennsylvania portions of the Upper Delaware Valley also were reviewed, including significant Phase II and Phase III reports for Pennsylvania and New York areas adjacent to the Upper Delaware Valley. Relevant published literature also was consulted for the project area in addition to select CRM reports and the published literature for the broader Middle Atlantic Region.

Radiocarbon dates from sites in the study area are compiled in Table 1. Dates are organized sequentially through time, starting with the earliest. Other biface types found in the same dated contexts also are listed. In most cases I have not challenged the original authors’ typological assessments. Radiocarbon dates are included for areas adjacent to the Upper Delaware as defined here. How far afield to extend the collection of dates was an arbitrary decision on my part in order to enhance the data base.

Table 1.

Radiocarbon dates related to triangular bifaces in the Upper Delaware Valley, New Jersey, Pennsylvania, and New York.

Biface type	Date BP	Calibrated*2 sigma age BC/AD	Calibrated* median age BC/AD	Site & context	Reference
Archaic Triangle	6340 ± 70 Beta 105802	5477–5207 BC	5326 BC	36Nm140, Oberly Island, associated with dated charcoal sample	Siegel et al. (1999: 40); Siegel et al. (2001: 27, 31, Table 1)
Triangular points (unacknowledged in original report))	4980 ± 110 I-6599	3992–3626 BC	3782 BC	28Wa2, Harry’s Farm, part of dated Zone 4	Kraft (1975: 23–25, Figure 18, Table 11); Stewart and Cavallo (1991: 24–25)
Lamoka, Brewerton, Beekman Triangle, Kittatinny	4510 ± 40 Beta 266910	3361–3090 BC	3214 BC	28Wa290, associated with dated deposit	Lee et al. (2010: 4–41 to 4–42, 6–9, 6–14, Appendix C)
Vestal Notched, Brewerton Side Notched, Vosburg, Beekman and other Triangles	4450 ± 130 I-4837	3520–2868 BC	3147 BC	10 Mile River Rockshelter, part of dated basal artifact deposit, Sullivan County, NY	Funk (1989: 47, Figure 6, Table 5); Funk et al. (1971)
Poplar Island, Macpherson, Lamoka-like, Beekman Triangle, Koens- Crispin/Snook Kill	3920 ± 95 I-6598	2673–2134 BC	2400 BC	28Wa2, Harry’s Farm, part of Zone 3 dated component	Kraft (1975:29, 33, 36, 38–39, Figures 20, 22, Table 11)
Jacks Reef Pentagonal, Levanna	2580 ± 80 Beta 52248	899–476 BC	689 BC	36Nm15, Padula, Feature 6	Doershuk (1994: 313, 323, Figure 14.17)
Levanna, Jacks Reef Corner-notched, Jacks Reef Pentagonal	1660 ± 100 DIC-249 1400 ± 55 DIC-262	205–592 AD 545–695 AD	383 AD 633 AD	Gardepe (Una 16-4), associated with dated Zone 3 in Locus 1, Delaware County, NY	Funk (1993: Table 16, 1998:395, 399, 402)
Vestal, Susquehanna, Triangle	1180 ± 90 Beta-no lab # reported	672–999 AD	842 AD	Otego Yard Site (NYSM 121), Feature 152/153, Otsego County, NY	Hartgen Archaeological Associates (1988b:48, 79, 101–102, Table 32)
Levanna	1180 ± 80 1160 ± 80 1150 ± 80 1020 ± 80 1000 ± 70 990 ± 60 Beta-no lab # reported	675–994 AD 758–1016 AD 759–1020 AD 863–1211 AD 892–1190 AD 948–1186 AD	842 AD 862 AD 872 AD 1016 AD 1042 AD 1058 AD	Ouleout Site (8W-32-7), Feature 18, Delaware County, NY	Hartgen ArchaeologicalAssociates (1989:26, 32–34, Plates 2–4, Appendix II)
Triangle	970 ± 120 Beta 62433	859–1272 AD	1067 AD	36Pi4, Unit 1, Level 2, Feature 1D	Wall and Botwick (1995: 150–151)
Levanna	870 ± 75 Dic-166	1026–1268 AD	1155 AD	Fortin Locus 2, Delaware County, NY, in same level as dated Feature 30	Funk (1993: Table 17)
Triangle	680 ± 50 DIC-1372	1259–1399 AD	1311 AD	28Sx5, Medwin North, Feature 14	Williams et al. (1982:23, Tables 2, 7)
Triangular	640 ± 120 Y-2474	1151–1493 AD	1333 AD	36Pi13A, Faucett, part of dated component	Kinsey (1972:194, 197, 464, 1975:28)
Levanna	600 ± 30 Beta 378839	1297–1373 AD 1377–1408 AD	1346 AD	Otsiningo Market Site (SUBi-3041), Feature 5, Broome County, NY	Miroff (2014:Table 8)
Triangular	550 ± 80 Y-2338	1276–1483 AD	1375 AD	36Pi25, Kutay, Feature 75	Kinsey (1972: 253, 255, Table 9, Figure 73g, h; Figure 75)
Madison	540 ± 30 Beta 332933 300 ± 30 Beta 309042	1316–1354 AD 1389–1436 AD 1490–1602 AD 1612–1654 AD	1403 AD 1563 AD	Chenango Point South (SUBi-2776), Feature 176, Broome County, NY	Miroff (2012:Table 4.3)
Triangular	420 ± 45 DIC-1214	1418–1524 AD 1558– 1631 AD	1474 AD	28Sx266, Medwin Knoll, Feature 9	Williams et al. (1982:40, Tables 5, 7); Puniello (1980:Table 2)
Triangle, Otter Creek-like	210 ± 40 Beta 216050	1635–1696 AD 1725–1814 AD 1917–1950 AD	1766 AD	36Nm204, Feature 1–07	Hornum et al. (2009:46; Appendix III)
Triangle	150 ± 60 Beta 42909	1663–1895 AD 1903–1950 AD	1798 AD	36Mr5, Smithfield Beach/Pardee, Feature 121	Hennessy (1992:303, Appendix A)

*Calibrated with Intcal13.14c data (Reimer et al., 2013) using Calib 7.10 (Stuiver et al., 2017; cf. Stuiver and Reimer 1993). Calibrated 2 sigma age ranges represent the greatest relative area under the probability distribution curve, generally areas ranging from .90–1.00.

Table 1 includes 26 dates related to contexts in which triangular points have been found. Of these 73% (n=19) fall within the late Middle Woodland, Late Woodland, and Contact periods. I use the following scheme as a general referent for intervals of time when not citing a specific range of radiocarbon dates:

Paleoindian (10,000 BC–8000 BC)

Early Archaic (8000 BC – 6500 BC)

Middle Archaic (6500 BC – 3000 BC)

Late Archaic and Transitional Archaic (3000 BC – 1000 BC)

Early Woodland (1000 BC – 500 BC)

Middle Woodland (500 BC – AD 800/900), and

Late Woodland (AD 800/900 to contact with Europeans)

Discussion

Regional context

Early-on in the Upper Delaware Valley, Leslie (1963: 71, 74) discussed what he termed Archaic Triangles in his analysis of biface types in Wayne County, Pennsylvania and Sullivan County, New York. He gives as the basic criteria for identifying Archaic triangles their relative thickness and cruder chipping relative to Late Woodland specimens. He notes that some are morphologically similar to the Levanna and Madison types attributed to Late Woodland times, but “there is less intergrading between the Madison points and the Archaic Triangles than in the case of the Levanna type” (Leslie, 1963: 82). In turn, Ritchie’s (1961, 1971) metrics and illustrations of Levanna and Madison points show considerable overlap between these Late Woodland types.

Leslie assumes that the Archaic Triangles relate to the Middle and Late Archaic periods but provides no contextual data for his interpretation. His Archaic Triangle type grades into what he provisionally named the Damascus Triangle (Leslie, 1963:78). These are described as characteristically long but of triangular shape, and relatively rare. He speculated, conceding a lack of evidence, that they might be of Middle Archaic age. In his assignment of age, Leslie seems to be drawing partially on Ritchie’s (1961) initial typology for New York projectile points, which included the Brewerton Eared Triangle, and the potential relationship of triangular forms with Laurentian or Brewerton assemblages. Figure 2 depicts Leslie’s examples of early triangular types.

Figure 2.

Archaic triangles (left two) and Damascus triangles (right two) from the Upper Delaware Valley. Modified from Leslie (1963:74, 78).

By 1971, Ritchie (1971: 121, 127) had added the Late Archaic Beekman and Squibnocket triangle types to his New York typology. Kinsey (1972: 439–443) acknowledges that some triangles may be Late Archaic in age, but presumes that most from the Upper Delaware Valley sites that he examined are affiliated with Late Woodland components. His description (Kinsey, 1972: 443) of what he labelled as Elongate Triangular points, considered to be Late Woodland in age, would subsume Leslie’s Damascus Triangles. Kraft (1975: Figure 18) includes un-notched triangular forms as part of the late Middle Archaic Kittatinny Complex dated to 4980 ± 110 BP at Harry's Farm (28Wa2) in the Upper Delaware (see Table 1; Kraft, 1975: 24, Figure 18; Stewart and Cavallo, 1991: 24–25). Funk (1975: xiii) notes that some variants of Kraft’s Kittatinny point type grade into Brewerton Eared triangle and Beekman triangle points. Both Leslie (1963:76) and Kinsey (1972: 406) recognized Brewerton Eared triangles on area sites but commented on their rarity.

In 1976 Hunterbrook triangles were defined on the basis of two specimens found in an Archaic context at the Hunter Brook Rockshelter in the Hudson Valley of New York (Wingerson and Wingerson, 1976). In excavations they occurred at 27 in below surface and beneath what were identified as Beekman triangles, Kittatinny, Vosburg, Brewerton, Lecroy, and Palmer-like points. No radiocarbon dates are available for the deposit. On the basis of the shelter’s stratigraphy and a comparative analysis involving other sites with triangles in stratigraphic contexts, the Wingersons interpreted the Hunterbrook triangle as a Middle Archaic form with technological or cultural links to Late Archaic types of triangles.

The two examples from the Hunter Brook Rockshelter are medium sized, nearly equilateral points with excurvate sides. Bases are ground and the points are noticeably bifacially thinned up to one third of the length of the point (Wingerson and Wingerson, 1976: 25). The points are about 2 cm long with widths of 2.0 and 2.5 cm (Wingerson and Wingerson, 1976: Figure 1). Drawing on triangles found in other Archaic contexts in the region, the Wingerson's (1976:26) defined the type on the basis of 24 specimens as: a medium sized, equilateral (80%) triangle with a concave base; bases moderately to heavily ground with bifacial thinning about 1/3 the length of the point on all specimens; 67% are 1 in – 1.75 in (2.54 – 3.175 cm) in length, 25% are .75 in – 1 in (1.9 – 2.54 cm), and 8% are 1.33 in (3.33 cm) in length.

In size range and general shape, the Hunterbrook triangles overlap descriptive data for the Late Woodland Levanna type which lacks basal grinding (Ritchie, 1971: 31). Morphological similarities with Late Archaic Beekman triangles, which often exhibit slight to moderate basal grinding, can also be noted (Ritchie, 1971: 121). Late Archaic Squibnocket triangles also share morphological similarities (Ritchie, 1971: 127). It is difficult to compare any of the forms with Leslie’s proposed Archaic forms from the Upper Delaware Valley.

Attention was refocused on the antiquity of triangular points in the Delaware Valley with the excavation of the deeply stratified Area D site (28Me1-D) within the Abbott Farm National Historical Landmark (AFNHL). A collection of 49 triangular points/fragments from these deposits are clearly Archaic in age and used over a period potentially beginning as early as 4000/4500 BC (Stewart, 1990a, 1990b, 1998; Stewart and Cavallo, 1991: 25–26, Figure 2; Wall et al., 1996a: 9–10, Table 2; Wall et al., 1996b). The triangle assemblage is derived from at least 10 different occupation levels that include artifact clusters on charcoal stained living floors, and shallow, basin-shaped hearth pits. A summary of the site’s stratigraphy and the bifaces recovered are fundamental to an understanding of triangular points found in pre-Late Woodland contexts in the Upper Delaware and broader Middle Atlantic and Northeast regions.

Table 2.

Radiocarbon dates associated with archaic triangular points at area d (28ME1-D) Abbott farm national historic landmark.

Context	Date BP	Calibrated*2 sigma age BC	Calibrated* median age BC
Base of IIC horizon, level 15	4310 ± 290 Beta 15186	3651–2192	2943
Top of IIIC horizon, level 17	4410 ± 110 Beta 15187	3373–2866	3093
IVC horizon, level 22	5450 ± 200 Beta 15185	4719–3907	4279
4’ below top of IIIC: Feature 55-2; Sample Cat No.277	5120 ± 120 Beta 37995 5040 ± 290 Beta 34007	4177–3661 4177–3626	3916 3839
6’ below top of IIIC, Feature 79	5320 ± 170 Beta 37996	4491–3765	4145

*Calibrated with Intcal13.14c data (Reimer et al. 2013) using Calib 7.10 (Stuiver et al. 2017; cf. Stuiver and Reimer 1993). Calibrated 2 sigma age ranges represent the greatest relative area under the probability distribution curve, generally areas ranging from .90 – 1.00.

Area D is one of several lowland sites located within the AFNHL at the Piedmont to Coastal Plain transition. Prior to highway construction, the site was flanked on three sides by fresh water and a tidal marsh. Table 2 organizes radiocarbon dates from the site. Figure 3 depicts an idealized profile of the deposits (compiled from Stewart, 1998; Wall et al., 1996b: Table 1, Figure 8). Figure 3 retains the original labelling of horizons. The IIIC, IVC, VC, VIIC, VIIIC and IXC horizons all include variable expressions of lamellae. In some cases the lamellae may be the result of pedogenesis and thus should be considered as B/C horizons. However, the possibility remains that some lamellae are depositional given their position below the current water table and the likelihood that the water table has fluctuated over the millennia as a result of sea level rise.

Figure 3.

Idealized profile of deposits at the Area D site, 28Me-1D. Modified from Stewart (1998) and Wall et al. (1996b: Figure 8).

Alluvial sediments and artifact deposits post-dating 4310 ± 290 BP are found in the upper 4 – 6 ft of the profile. Triangular points occur in the uppermost portion of the profile with Fishtails, Bare Island-like, Broadspear, and Poplar Island-like bifaces derive from progressively deeper excavation levels prior to the reappearance of triangles. The 4310 ± 290 BP date is derived from the base of the IIC horizon. A date of 4410 ± 110 BP is from the top of the IIIC horizon. The IVC horizon is associated with an assay of 5450 ± 200 BP. These three dates appear reasonable given their stratigraphic positions. The latest point of overlap between the calibrated dates from the IIC and IIIC horizons is 2866 BC; statistically the two dates could represent the same relative point in time. There is no overlap with the date from the IVC horizon with the latest portion of its calibrated range at 3907 BC.

Radiocarbon dates from successively deeper excavation levels, situated well below the water table, seem problematic at first glance. The two dates derived from Feature 55-2 and Sample No. 277 are statistically alike. Although their probability ranges overlap that of the date from the IVC horizon, their considerable stratigraphic separation from the IVC horizon indicate that they represent distinctive ages. The calibrated date from Feature 79 also overlaps that of the three dates situated stratigraphically above it. But again, its relative position indicates that it represents a distinctive time, as it can considering the assay’s calibrated range.

Recognizing that a single assay is a probability statement representing what was once a single moment in time (i.e., the death of the organic sample being tested), points within the calibrated range of each of the dates could be selected to create a coherent sequence that becomes increasingly older with depth. Of course, this would assume that the dated samples themselves are without problems which may not be the case given their provenience in a waterlogged matrix. Accepting such assumptions, one could argue that the level at which Feature 79 occurs pre-dates 4000 BC.

Select attributes for the Area D triangles are summarized in Table 3. More detailed compilations of attributes are found in Katz (2000) and Stewart (1998). Examples are depicted in Figure 4 and are arranged stratigraphically. Traits common to Archaic triangles at Area D include: a preference for isosceles shapes, although other forms are represented; a preference for chert as a raw material; absence of grinding on any lateral margins; lateral margins that are typically straight to slightly excurvate; bases which are often straight but range from slightly convex to slightly concave. There are fairly common ranges and trends in length, width, and other metrical attributes. Attributes do not appear to change significantly though time although basal width seems to increase slightly with depth/age. The morphology of the triangles from the Late Woodland deposits on-site is readily encompassed by the variation evident in the assemblage of Archaic forms (Wall et al., 1996a: Table 2).

Table 3.

Select attributes of archaic triangles, 28ME1-D.

Context	Sample size	Axial length cm RangeAverage	Basal width cmRangeAverage	Maximum thickness cmRangeAverage	Comments
IIC base	5, 4 generally whole	2.09–3.74 2.54	1.8–2.32 1.9	0.4–0.88 0.55	Sample includes isosceles and equilateral forms. 50% reveal patterned treatment of base – short, steep pressure flakes that are continuous along one face of the base, and long, deep pressure flakes centrally focused on the opposing face of the base. No grinding apparent on any margin. Lateral margins are straight to slightly excurvate; bases are predominantly straight but range from slightly convex to slightly concave.
IIIC	8, 7 generally whole	1.8–4.73 3.14	1.69–3.1 2.09	0.4–0.87 0.56	75% are isosceles triangles, remainder are equilateral and pentagonal. 63% reveal patterned treatment of base as described above. No grinding apparent on any margin. Lateral margins are straight to slightly excurvate; bases are predominantly straight but range from slightly convex to slightly concave.
IVC	6, 5 generally whole	1.8–2.7 2.38	1.7–2.0 1.8	0.3–0.75 0.33	80% are isosceles triangles, remainder are equilateral. 20% reveal patterned treatment of base as described above. No grinding apparent on any margin. Lateral margins are straight to slightly excurvate; bases are straight to slightly convex.
VC	10, 7 generally whole	1.8–2.9 2.5	1.72–2.6 1.98	0.31–0.8 0.44	90% are isosceles triangles, remainder are equilateral. 77% reveal patterned treatment of base as described above. No grinding apparent on any margin. Lateral margins are straight to slightly excurvate; bases are predominantly straight but range from slightly convex to slightly concave.
VIC	7, 6 generally whole	2.1–3.8 2.82	–	0.4–0.64 0.49	66% are isosceles triangles, remainder are equilateral. 50% reveal patterned treatment of base as described above. No grinding apparent on any margin. Lateral margins are straight to slightly excurvate; bases are predominantly straight to slightly concave.
VIIC	1 fragment	NA	NA	NA	NA
IXC	2, 1 generally whole	2.2–3.02 2.61	2.44	0.43–0.45 0.44	100% are isosceles triangles. 50% reveal patterned treatment of base as described above. No grinding apparent on any margin. Lateral margins are slightly excurvate; bases are slightly convex to slightly concave.
XC	4, 1 generally whole	3.0	2.6	0.32–0.6 0.45	Whole specimen is an isosceles triangle. Whole specimen reveals patterned treatment of base as described above. No grinding apparent on any margin. Lateral margins are slightly excurvate. Base is concave.
XIC	3, all generally whole	2.6–2.92 2.97	2.01–2.65 2.34	0.4–0.51 0.47	100% are isosceles triangles. None reveal patterned treatment of base as described above. No grinding apparent on any margin. Lateral margins are straight to slightly excurvate; bases are slightly concave.
XIIB	1 fragment	NA	NA	NA	NA

N.B. Two additional whole triangles were found in deep deposits but lack exact provenience.

Figure 4.

Examples of Archaic triangles from 28Me1-D. Top row from base of IIC horizon with Teardrop and Kittatinny points; 2nd row from IIIC horizon; 3rd row from IVC horizon; 4th row from VC horizon; 5th row from VIC horizon; 6th row left-IXC horizon, right-XC horizon; 7th row from XIC horizon. Photo courtesy of Rob Tucher.

In what I originally believed was a distinctive basal treatment, short, steep pressure flakes are continuously removed along one face of the basal margin. On the opposite face, longer and deeper pressure flakes are removed, and clustered on central portions of the margin (i.e., they are not continuous across the entire margin). This strategy results in one face of the basal margin having a steeper angle than the other. In some cases, this asymmetrical quality of the angle of the basal margin is achieved using only short pressure flakes, which is why I believed that the asymmetry is a distinctive trait that can be considered on its own, as well as in combination with the patterned removal of pressure flakes.

Katz (2000) analyzed and compared assemblages of Archaic triangles from Area D with those of Archaic age from three sites in the vicinity of Liverpool, Pennsylvania in the Susquehanna Valley: 36Pe16, 36Pe60 and 36Pe61. On the Susquehanna Valley sites the triangles were part of Broadspear/Transitional Archaic components (Miller et al., 2009). The Archaic point assemblages were then compared with an assemblage of Late Woodland triangles from the Gropp’s Lake site located adjacent to the AFNHL (Stewart, 1987). Katz concludes that the metric and morphological attributes that are commonly encountered in type descriptions (i.e., length, width, thickness, and edge curvature) show notable overlap between Archaic and Late Woodland triangles, with basal width being a possible exception (Katz, 2000:81). Earlier, Kotcho (1998) used length, width and thickness in a discriminant analysis attempting to distinguish Area D Archaic triangles from Late Woodland specimens. The samples were metrically quite similar and could not be statistically differentiated.

Katz’s (2000) analysis of Archaic triangles and comparisons with Late Woodland triangles from Gropp’s Lake and other published Late Woodland data (e.g., Custer, 1983) does not support the basal treatment described above as a potentially unique feature of Archaic triangles. Katz’s study, however, did identify some characteristics of the Archaic triangles that set them apart statistically from those of the Late Woodland period. In comparison with the Late Woodland triangles from Gropp’s Archaic triangles have (Katz, 2000: 80–81):

-a greater proportion of straight-based forms;

-a smaller mean basal depth;

-smaller mean basal widths;

-a smaller proportion of bases exhibiting basal grinding;

-a smaller mean edge angle; and

-a greater proportion of excurvate lateral margins.

While the above distinctions are significant they are not presence/absence criteria that definitively can allow an analyst to identify any individual specimen as Archaic versus Woodland, especially when dealing with surface assemblages of artifacts. In conjunction with other data, the noted attributes increase the likelihood of identifying early triangles from buried but undated contexts.

The results of Katz’s functional analysis (breakage patterns, edge damage, use wear) of Archaic and Late Woodland triangular points is summarized in Table 4. Both populations functioned as projectiles and cutting implements. Although Late Woodland points provided more evidence of use as projectiles, the differences were not statistically significant (Katz, 2000: 89).

Table 4.

Functional comparison of late woodland triangular points with archaic assemblages from 28ME1-D, 36PE16, 36PE60, and 36PE61 (source: Katz, 2000: Table 6.1).

	Gropp’s lake		Area D		Liverpool
Function	No.	%	No.	%	No.	%
Knife function only	17	39.5	11	55	9	50
Projectile function only	13	30.2	5	25	2	11.1
Knife and projectile uses	9	20.9	1	5	5	27.8
Drill or perforator only	4	9.3	2	10	1	5.6
Drill and knife	0	0	1	5	1	5.6
Knife total	26	60.5	12	60	15	83.3
Projectile total	22	51.2	6	30	7	38.9
Drill total	4	9.3	3	15	2	11.1

Archaic triangles are known from other sites in adjacent areas. At the Gropp's Lake site (28Me100G) triangular points, found in lower excavations levels of stratified deposits, were dismissed originally as probable intrusions from higher strata (Stewart, 1987: Figure 5.7). The deep triangles occur stratigraphically just above Stanly/Neville and Brewerton Notched points (Stewart, 1987: Plate 6.25). The deep triangles, Stanly/Neville, and Brewerton points are found in a soil horizon which pre-dates 2420 ± 100 BC, a radiocarbon assay from the base of the overlying stratum (Stewart, 1987: V-4).

At the White Horse West site, 28Me119, also adjacent to the AFNHL, triangular points identical to those from Area D occur in the deepest Archaic context along with Kanawha, Stanly/Neville, and contracting stemmed points which could represent variants of the Stark or Morrow Mountain types (McLearen and Fokken, 1986: Plate 6.2). These artifacts are mixed with items more typical of the Late Archaic period. The context has not been radiocarbon dated.

Approximately 22 miles (36 km) to the east of the AFNHL, five radiocarbon dates (Cavallo, 1981:8) and trianguloid bifaces/points assigned to the late Paleoindian period from the Turkey Swamp site, 28Mo305, represent an equivocal situation. The earliest assay, 8739 ± 165 BP, when calibrated at two sigmas (Calib 7.0; Reimer et al., 2013) could feasibly indicate a very late Paleoindian occupation ca. 8258 BC. However, the calibrated ranges of all of the dates crosscut the chronological boundaries typically associated with the Early and Middle Archaic periods. The radiocarbon dates and stratigraphic context could represent a Middle Archaic context. Lithic preferences, the use of cobble and pebble sources of chert, and lithic reduction strategies mimic those evident in the Archaic triangle assemblages at Area D (contra Stewart and Cavallo, 1991:24).

Mounier (2008:187) recognized Archaic and Late Woodland triangles in excavations of Locus A at 28Bu407. Triangular points morphologically similar to the Area D assemblages have been excavated from the deepest levels of 28Gl210, another coastal locality situated about 36 miles (58 km) to the southwest of the AFNHL (Louis Berger and Associates, Inc., 1992:Plates 9.2, 9.3; Lothrop and Koldehof, 1994:107–113). They are minimally Late Archaic in age being situated stratigraphically below a Lackawaxen/Poplar Island component. The deposits have not been dated. Also located in Gloucester County, four triangles were found in a level with a variety of stemmed points attributable to the Late Archaic period at 28GL228. A thin sterile layer of sand separated the Late Archaic deposit from an overlying Jacks Reef component at the site (Bello et al., 1998).

Farther south in buried deposits at Locus 5 of 28Sa214 a small triangular point was positioned stratigraphically below a Bare Island point, which in turn was located below a Poplar Island point (Heinrich and Hinshaw, 2017). Subsequent excavations at Locus 5 indicate that the context of the triangle find post-dates 6002 – 6271 BP, a 2-sigma calibrated radiocarbon date from an underlying 0.25 ft excavation level (Heinrich, 2018: 5–40, 5–42, Figure 5.17). Jacks Reef points and triangles are associated in a Middle Woodland component at 28Gl170 (Mounier, 2003:216). The association of triangles and Jacks Reef points is not unexpected in regional deposits given the long-recognized overlap of their use histories (cf. Custer, 2001:34; Ritchie, 1971: 26; Stewart, 2018: Table 2)

Farther afield in the Coastal Plain of Maryland, triangular points occur in stratified deposits at the Pig Point site, 18An50, and are attributed to both the Late Archaic and Early Woodland periods (Luckenbach et al., 2010).

The size range of Late Archaic Beekman triangular points is replicated in the Area D collection. Equilateral forms predominate, and the morphology of lateral and basal margins encompasses the variability seen in the Area D specimens. However, Beekman triangles typically reveal ground bases, something not seen in the Area D artifacts. Basal grinding also distinguishes Hunter Brook triangles from those at Area D, although morphologies are comparable. The basic Hunter Brook, Beekman, and Squibnocket morphologies overlap entirely with the Archaic triangles from Area D and the Susquehanna Valley sites (Katz, 2000:98).

Beekman triangles from New York and New England range in date from approximately 2500 BC - 2800 BC, and are considered to be a part of the Vosburg Complex (Ritchie, 1971: 121; Ritchie and Funk, 1973:341). Squibnocket triangles are associated with later phases of the Late Archaic in eastern New York and southern New England, and seem to be an established type by 2200 BC (Ritchie, 1971: 127–128; Ritchie and Funk, 1973:341–342; Snow, 1980:223–228). As noted by Johnson et al. (1984: 27), “The morphological overlap of Late Archaic Beekman and Squibnocket Triangles is so great that they have been grouped together under the new type name - Small Triangle.” Small Triangles range in age between roughly 3000 BC and 1000 BC in New England (Johnson et al., 1984: 98–99). Doucette (2005: 26–27, Figure 5) defined a new type of triangular point (Snappit) on the basis of 37 specimens from excavations at Annasnappet Pond in Massachusetts. Most (over 62%) have a concave ground base and slightly serrated edges. They are not as short and equilateral as the Squibnocket type and are not as wide and with as straight lateral margins as the Levanna type of Late Woodland times. Possible similarities with Beekman triangles are noted. The triangles are associated with three features with dates of 5100 ± 40 BP, 5810 ± 40 BP, and 7210 ± 70 BP (Doucette, 2005:27, Table 1). In general morphology they are similar to the Archaic triangles at Area D, but are distinguished by the incidence of basal grinding and the slight serration of some lateral margins.

Figure 5.

Archaic triangles from Oberly Island, 36Nm140. Modified from Siegel et al. (1999: Plate 63b-d).

Additional evidence exists for Archaic triangles in the Susquehanna Valley of Pennsylvania supplementing the assemblages from the sites involved in Katz’s (2000) analysis. On Calver Island (36Da89) triangles are associated with deposits dated to ca. 4300 BP (Miller et al., 2010: 94–95). At the Skvarek site (36Lu132) a triangular point with straight base and basal grinding is associated with a date of 4160 ± 70 BP (Miller, 1994; Miller, 1998:116). General similarities with the Hunter Brook type are noted. With the exception of basal grinding the point’s morphology falls within the variation represented by the Area D assemblages. The triangle occurs in the same stratum as a Lackawaxen point (Miller 1998: 107). A single basally thinned triangle is associated with three bifurcated-base points and a date of 7390 ± 110 BP in Segment A of the Middle Archaic deposits at the West Water Street site, 36Cn175 (Custer, 2001: 84; Custer et al., 1994, 1996:30, 33, Figure 22, 23).

Component IV at the nearby Memorial Park site (36Cn164) included Brewerton Side Notched, Brewerton Eared Notched, Brewerton Corner Notched, Chillesqueque Triangle, and Vosburg point types (Cremeens and Hart, 2009: 58). Dates of 6355 ± 155 BP and 6115 ± 265 BP are associated. Beekman triangles are part of Component V, along with Otter Creek, Brewerton Side Notched, and Brewerton Eared points. Dates of 5830 ± 130 BP and 5790 ± 240 BP are associated with the deposit (Cremeens and Hart, 2009: 58, 60). In their review of the work of East et al. (2002) at the East Bank site, Bergman et al. (2020) note that Archaic triangles are found between 6260 ± 40 BP and 3620 ± 60 BP.

About 15 miles to the north of the Delaware Basin in Otsego County, New York, isosceles triangular points from the Camelot No.2 site are associated with a date of 4795 ± 230 BP (Funk, 1993: Figure 1; 1998: 216, Plate 64–figures 33, 34). The points were originally described as trianguloid resembling the Brewerton Eared Triangle type (Funk, 1988:27). In later publications they are simply identified as Brewerton Eared Triangles. At 36Ly290 a triangle is associated with Feature 1. Dates for the feature are 5590 ± 70 BP and 5580 ± 70 BP. Triangles also appear in contexts of Middle Woodland age at the site (Reinbold et al., 2000).

The study area

Triangles associated with Archaic-aged dates (Table 1) are known for the project area and mimic those reported for the broader region. From earliest to latest they are: 6340 ± 70 BP, 4510 ± 40 BP, 4980 ± 110, 4450 ± 130 BP, and 3920 ± 95 BP. Calibrated medians range from 5326 BC – 2400 BC. The earliest date, 6340 ± 70 BP, relates to deposits at Oberly Island (36Nm140) with dated material in close association with one of three Archaic triangles found in the lower portions of the upper Bt horizon in the stratigraphic sequence (Figure 5; Siegel et al., 2001:31). No type designation was assigned. Length, width and thickness metrics include a mean length of 2.54 cm, mean width of 1.85 cm, and a mean thickness of 0.45 cm (Siegel et al., 2001: Tables 5, 11); they fall within the ranges of the Area D assemblage.

Table 5.

Range of regional radiocarbon dates for biface types associated with triangles in the study area.

Biface type	Range of dates* BP	References
Brewerton Side Notched	6115 ± 265 to 3960 ± 100	Anderson et al. (2000: 6–218 to 6–219); Custer (2001: Figure 11); Inashima (2008: 201; 2011: 97); Johnson (2002: 2–41); Stewart (2018:Table 2)
Brewerton Corner Notched	6115 ± 265 to 4300 ± 180	Anderson et al. (2000: 6–217); Custer (2001: Figure 11); Inashima (2008:200; 2011:96)
Lamoka	5383 ± 250 to 2650 ± 150	Custer (2001: Figure 11); Funk (1993:Table 17); Inashima (2008: 222–223; 2011:108–109); Johnson (2002:2–44 to 2–45); Justice (1995:129); Ritchie (1965: Figure 1); Ritchie and Funk (1973: Figure 1); Stewart (1987: 170, Plates 21–23, Appendix 1); Wall (2015:40–47)
Brewerton Eared Triangle	5180 ± 200 to 4220 ± 160	Custer (2001: Figure 11); Funk (1993: Table 17); Inashima (2008: 201); Justice (1995: 123)
Vosburg	6115 ± 265 to 3965 ± 155	Custer (2001: Figure 11; Funk (1976: 243); Inashima (2011:127–128); Justice (1995:116); Kinsey (1972, 1975:61, Figure 23)
Macpherson	4160 ± 140 to 3660 ± 120	Funk (1965:137); Kinsey (1972:413); Kraft (1975:36–40)
Poplar Island	4470 ± 50 to 2080 ± 90	Custer (1996:139–145; 2001:43, 90–113); Inashima (2008:245); Miller et al. (2010:93, Table 1); Stewart (2018:Tables, 2, 8)
Vestal	4710 ± 40 to 3775 ± 115	Funk (1993: Table 17; 1998: 53, 451, Plate 157); Funk and Wellman (1984); Knapp (2011: Tables 4.3, 4.4, Appendix 3); Kudrle (2005:45, Table 14, Appendix 6)
Jacks Reef Pentagonal	1840 ± 90 to 1000 ± 45	Custer et al. (1990:161, Table 2); Funk (1993:Table 17, 1998:Plate 43); Inashima (2008: 217); Johnson (2002:2–60 to 2–61); Lowery (2013); McConaughy (2013); Rieth (2013:Table 1); Walker (2013)
Jacks Reef Corner Notched	1660 ± 100 to 1020 ± 100	Anderson et al. (2000:6–239 to 6–241); Custer et al. (1990); Funk (1993:Table 17); Inashima (2008: 216–217; 2011:107); Johnson (2002:2–60 to 2–61); Lowery (2013); McConaughy (2013); Walker (2013)

*Ranges shown consist of the earliest and latest radiocarbon dates not clearly discounted by the investigators.

In other cases triangles are described as the Beekman type and co-occur with a number of other point types that chronologically straddle the Middle and Late Archaic periods. At 28Wa290 and Harry’s Farm (28Wa2) they are found with Kittatinny points, a possible association seen late in the Archaic sequence at Area D (Figure 4, upper row of bifaces). Although illustrated as part of Kraft’s (1975: Figure 18) Kittatinny Complex at the site, triangles are not directly acknowledged and metrical data are not provided. The most obvious example (Kraft, 1975: Figure 18m) is an isosceles form approximately 3.2 cm in length with a width of 2.5 cm (scaled measurements from photograph). Kittatinny points are presumed to be a late Middle Archaic form, but have only been dated twice in the Upper Delaware Valley (Stewart, 2018: Table 2). The single date from 28Wa2 (4980 ± 110 BP) where the type was defined remains the earliest. A date from 28Wa290 (4510 ± 40 BP) overlaps the two dates from Area D where the type is associated with Archaic triangles. A Beekman Triangle also occurs in the same dated context as the Kittatinny point at 28Wa290. Kittatinny points occur in three different Middlesex Complex/Early Woodland burials at the Rosenkrans site (Kraft, 1976: 16, 25, 29, Figure 2, 10, 12). In one case they co-occur with Vosburg, Lackawaxen, and Lehigh/Susquehanna Broadspear points (Kraft, 1976: 25, Figure 10; Table 2 this report). Kraft believes that these older points were collected by mourners and deliberately included as grave goods along with caches of Early Woodland biface types that are contemporaneous with the internments.

Funk (1975: xiii) points out that some Kittatinny specimens grade into the Brewerton Eared Triangle and Beekman Triangle types. Leslie’s (1963:76) illustration of a Brewerton Eared Triangle from the Upper Delaware resembles the morphology of the Kittatinny point type described later by Kraft. At the Peake site (Delaware County, New York) a Brewerton Eared Triangle occurs in an excavation level immediately below a context dated to 4310 ± 90 BP (Table 1; Hartgen Archaeological Associates 1988a:52, Table 18). A triangular biface could be transformed with minimal effort into a Kittatinny point.

Comments made concerning the triangles found in the dated basal levels of Stratum 3 of the Ten Mile Rockshelter, Sullivan County, New York (Table 1) reiterate the difficulties in clearly distinguishing Late Woodland forms from those of Archaic age.

Of great interest in this context (basal levels of Stratum 3) are three triangular points. A thin, well-chipped, nearly equilateral specimen with straight sides and base (Figure 29) could easily pass for a Levanna point, except for the lightly but evenly rubbed base. A thin isosceles triangle (Figure 30) which lacks rubbing seems indistinguishable from the Madison type, a Late Woodland form. The third point (Figure 31) is convex-sided and concave-based and lacks basal rubbing, but matches very well the form of the Beekman Triangle type (Funk et al., 1971: 36).

These comments again make clear that the attributes of all triangles in a single context both overlap and vary from those employed in existing typological descriptions of Archaic and Late Woodland forms. Figure 6 depicts the early triangles from the rockshelter.

Figure 6.

Triangular points from the base of Stratum 3 at the Ten Mile Rockshelter, Sullivan County, New York. Modified from Funk (1989: Figure 6) and Funk et al. (1971: Plate VIII).

Triangles appear in a number of suggestive, but not directly dated, early contexts. At 36Pi239 situated in small floodplain along the Lackawaxen River, what is described as an Egypt Mills point was found in a buried A horizon (Stratum IV) which also included Historic period artifacts. A triangular point which the investigators typed as a Late Woodland Madison form was found in the associated B horizon, Stratum V (Snyder and Petyk, 2010:95, Plate C4). If the stratigraphy is intact the triangle likely pre-dates the Late Woodland period.

Taking a Middle Atlantic region perspective, Table 5 summarizes the range of radiocarbon dates associated with biface types that co-occur in contexts with triangles in the Upper Delaware study area. A comparison of these ranges with the associations noted in Table 1 presents no clear contradictions. One gets the impression from reading CRM reports that the Brewerton Side and Corner Notched types are the default forms used to characterize any moderately-sized biface found on surface sites and in contexts that might be Late Archaic in age (Custer, 2001:66). Illustrations of points ascribed to Brewerton assemblages certainly include a wide range of forms (e.g., Ritchie, 1971: Plates 4, 7). While acknowledging that the Brewerton Phase is not well understood in the Upper Susquehanna and Upper Delaware valleys, Hohman and Versaggi (2003:7) suggest that the points in Brewerton series may simply be a variant of generic bifaces found on Late Archaic sites. Although dates for Vestal in the Upper Susquehanna range from 2400 BC – 1800 BC, Funk (1993:192–194) sets aside early dates and assigns Vestal bifaces to the time 1900 BC – 1800 BC or possibly 1600 BC. In part his dissatisfaction with these earlier dates stems from their overlap with the use history of Lamoka bifaces; stratigraphic relationships between Lamoka and Vestal components seen repeatedly at sites in the Upper Susquehanna Valley argue against their contemporaneity and for the chronological priority of the Lamoka type (Funk, 1993:193–194). Still, radiocarbon dates indicate the overlap of their chronologies even assuming a 1900 BC – 1800 BC time frame for Vestal. Ritchie (1971: 130) notes their definite association at the Castle Garden site in Broome County, New York. The admixture of types in dated Feature 168 at the Chenango Point site (Knapp, 2011:Tables 4.3, 4.4, Appendix 3; Stewart, 2018:Table 2) may relate to site formation processes and not represent a contemporaneous association.

The deepest 6 in level (#5) of the Friedman II site (28Sx16) contained Beekman triangles with an extensive array of other point types generally associated with the Middle Archaic, Late and Transitional Archaic periods. Higher in the stratigraphic sequence, Level 3 also included Beekman triangles along with Late and Transitional Archaic point types (Kinsey, 1972: 334–335).

In excavations on the southern end of the Manna site (36Pi4) the base of a triangular biface derives from Level 12 at 4.85 ft – 5.05 ft (147 – 153 cm) below datum. Level 16, at 5.75 ft- 6.2 ft (175 – 188 cm) below datum, is associated with a radiocarbon date of 4550 ± 180 BP (Stewart et al., 2015: 69–71; Wall and Botwick, 1995:155). A Late Woodland feature occurs at the base of the plowzone in Level 2 (1.85 ft – 2.0 ft; 45 – 60 cm) and is dated at 970 ± 120 BP and a triangular point was found in Level 3 at 2.15 ft– 2.2 ft or 65 – 67 cm below datum (Stewart et al., 2015: 68, Table 7; Wall and Botwick, 1995: 150–151).

A large isosceles triangle with an approximate length of 5 cm and width of 2.8 cm was found at the Rosenkrans site (28Sx2) in Burial 10 attributed to the Middlesex Complex of the Early Woodland period (Kraft, 1976: Figure 15). The point is not identified in the key to the illustration of Burial 10 artifacts or in the narrative text (Kraft, 1976: 38). It is part of a projectile cache Kraft (1976: 38) notes is “very close to Burial No. 10, but possibly in a separate pit”. In the case of other burials Kraft (1976: 16, 25, 29, Figure 2, 10, 12) identifies associated points whose original date of manufacture and use predate the internments. Dates for related burials on-site are 2560 ± 120 BP and 2400 ± 60 BP (Kraft, 1976: 23, 31).

A triangle occurs in Stratum VIII, Level 9 at Smithfield Beach (36Mr5) in a deposit attributed to a late Transitional Archaic/Early Woodland context (Hennessy, 1992). What is described as a Levanna triangular point was found in Stratum IV at Sandts Eddy (36Nm12) and presumed to be Late Woodland in age (Bergman et al., 1996a:58). However, the context appears to relate primarily to the Transitional Archaic and Early to Middle Woodland periods. Feature 6 at the Padula site (36Nm15) is a dump of fire cracked rock. Mixed in with the fire cracked rock were points identified as a Jacks Reef pentagonal and a Levanna triangle. In contrast a radiocarbon date of 2580 ± 80 BP associated with the feature (Doershuk, 1994:313, 323, Figure 14.17) would date the context to the Early Woodland period.

Current excavations at the Sibum site by Donald Kline in the Delaware Water Gap (Monroe County, Pennsylvania) have recovered triangular points in a stratified sequence (Figure 7) involving what appear to be four distinct soil sequa. Analysis of the recoveries from two, 10 ft x 10 ft blocks is ongoing and no radiocarbon dates have yet been obtained. For a perspective in evaluating the sequence, pottery, triangles, Fox Creek and other stemmed points occur in the plowzone. Fishtail points are recognized in the 16 – 19 in and 19 – 22 in levels. Fishtail and Normanskill points are found in the 22 – 25 in level, and Poplar Island, Lamoka-like, stemmed and notched points in the level from 25 – 28 in. Lecroy and notched points are in the 28 – 31 in excavation level, and corner and side notched points are found at 31 – 34 in. What seem to be the most recognizable Early/Middle Archaic biface types are depicted in Figure 8 for comparison with the sequence of triangular points.

Figure 7.

Triangular points from the stratified deposits at the Sibum site, Minisink Hills, Pennsylvania.

Figure 8.

Probable Early/Middle Archaic points from the Sibum site with below datum depths. Left to right: Lecroy, Kanawha Stemmed, untyped bifurcate-like, Kirk Corner-Notched, MacCorkle, Abbott.

Defining occupation surfaces and discrete chrono-stratigraphic units at the site has not been completed so it is premature to speculate about how long a particular occupation surface was stable, or the amount of time represented by individual chrono-stratigraphic units. Therefore the contemporaneity of biface types in a given excavation level cannot be assumed. Nonetheless, the vertical distribution of biface types and gross correlations of excavation units with depositional episodes and soil horizons suggest that triangles minimally relate to the Late and Transitional Archaic periods with some of potential Middle Archaic age.

Descriptive measurements and observations for the Sibum triangles are tabulated in Table 6. Asymmetrical thinning refers to a series of short shallow pressure flakes on one face of the biface’s base with deeper and longer pressure flakes on the opposite face, the treatment once thought to be a distinctive feature of early triangles. The only noticeable difference between the bifaces in the plowzone and those of lower levels are the basal concavities of the plowzone finds. There is considerable overlap in all other features replicating the conclusions of Katz’s (2000) controlled study. The elongate, isosceles forms at 19 – 22 in and 31 – 34 in resemble those seen in the Area D assemblage.

Table 6.

Descriptive data for triangular points from the Sibum Site Minisink Hills, Monroe County, Pennsylvania.

Context	Sample size	Axial length cm	Basal width cm	Maximum tickness cm	Comments
Plowzone	7, 4 relatively whole	4.5	5.1	1.3	The largest specimen appears to be a manufacturing failure. Basal grinding on 2nd specimen listed. Base of 3rd and 4th specimens listed asymmetrically thinned. All equilateral.
		3.6	3.1	0.6
		3.0	2.6	0.6
		2.7	2.5	0.6
16–19” below datum	4 relatively whole	3.45	1.9	0.8	Largest specimen an unfinished, late manufacturing stage. Base of 4th specimen listed asymmetrically thinned. Three equilateral, one isosceles.
		3.0	3.2	0.8
		2.9	2.3	0.4
		2.0	2.2	0.45
19–22” below datum	3 relatively whole	8.1	2.5	0.9	Basal grinding on 3rd specimen listed. One equilateral, two isosceles.
		5.1	2.4	0.6
		3.0	2.5	0.4
25–28” below datum	2, 1 relatively whole	3.9	3.0	0.95	Both appear to be unfinished, late manufacturing stages. Whole specimen nearly equilateral.
28–31” below datum	4, 2 whole	2.6	2.3	0.4	Base of 2nd specimen listed asymmetrically thinned. All equilateral.
28–31” below datum	4, 2 whole	2.1	1.8	0.4
31–34” below datum	1 relatively whole	4.0	2.5	0.9	Basal grinding. Isosceles.

Triangular points are typically assigned to the late Middle Woodland and Late Woodland periods, continuing in use during historic times and encounters with Europeans (e.g., Custer, 2001:113; Ritchie, 1971: 31, 33) and presumed to herald the widespread use of the bow and arrow. Previous discussions demonstrate that the morphology and function of triangles from this, and earlier time frames show considerable overlap. Triangular points from the late Middle Woodland and Late Woodland periods are the best dated in the study area. The 2580 ± 80 BP date from the Padula site (36Nm15) is considered to be in error. Associations of Vestal, Susquehanna, and triangular bifaces in a dated late Middle Woodland/LateWoodland context at the Otega Yard site, Otsego County, New York also is considered to be spurious. The co-occurrence of triangular and Otter Creek-like bifaces in a historic context at 36Nm204 is likewise problematic.

The largest portion of the late dates are associated with points identified as Levanna and range from 1660 ± 100 BP – 600 ± 30 BP. Only two dates 540 ± 30 BP and 300 ± 30 BP) are associated with the Madison type. Investigators chose not to type triangular bifaces in seven of the cases associated with radiocarbon dates ranging from 970 ± 120 BP – 150 ± 60 BP. The Levanna form is thought to pre-date Madison which Ritchie (1971: 34) states is the distinctive Iroquoian form in the Northeast. The date ranges for the two types can partially overlap, but what is emphasized is the gradual reduction in point size through time (Custer, 2001:113; Engelbrecht, 2014:353; Ritchie, 1971: 33–34).

Both the trend in size reduction and the affiliation of specific triangular point types with ethnic groups or archaeological cultures have effectively been challenged (Custer, 1983, 2001: 113–114; Stewart et al., 2000:VI-8, VII-26 to VII-30). In the greater Eastern Woodlands, the chronology of the two types is relatively identical (Justice, 1995: 227–228). While radiocarbon dates from the study area do not reflect this synchrony, contextual data do. For example, Levanna and Madison triangular point types are found in pit features that include European made goods at the Bell-Browning site, 28Sx19. Dated kaolin pipes ranging from 1620 – 1650 AD are associated (Marchiando, 1972:157). Marchiando argues that the pits were deep enough so as to not be disturbed by plowing or topsoil removal; therefore, the association of the non-native pipe fragments is a good one.

In a study of Madison triangles from an Iroquoian site Engelbrecht (2015:762) used T-tests to show significant differences between whole and broken and refitted points. “This calls into question the representativeness of existing point typologies” (Engelbrecht, 2015:765). However, he does not use these data to question the validity of typing triangular points to begin with, merely suggesting that more representative populations of artifacts need to be taken into account. Engelbrecht explains the differences in the large assemblage that he used in his analysis as potentially reflecting the difference between arrows/points used for hunting and those used for warfare. There is an overlap in the range of metrical data for Engelbrecht’s Madison assemblage with the metrical data (length, thickness) from Ritchie’s (1961, 1971: 31) definition of the Levanna type, although average metrics are generally distinctive.

Evidence indicating that Late Woodland triangles were used both as projectiles and tools in other activities was presented in previous discussions of early triangles. Supplementing this evidence are residue and microwear analysis from study area sites. At 36Mr133 one Late Woodland Levanna triangle (FS#131) tested positive to chicken antiserum which could be indicative of quail, turkey, grouse, and all gallinaceous fowl. It also tested positive for the Asteraceae plant family which could represent rabbitbrush, sunflower, or thistle. A second triangle (FS#185) also tested positive for chicken (quail, turkey, grouse, and all gallinaceous fowl) as well as pine and grasses (Hornum et al., 2009:76–79; Parr, 2006). Both points are illustrated in Figure 9. Five Levanna points from Sandts Eddy (36Nm12) served as projectiles and in light butchery according to microwear analysis (Kimball, 1996:Table 12.3). Microwear on two Madison points (artifacts 1740.16, 572.2) from Treichlers Bridge (36Nm142) revealed their use in bone and hide working (Church, 2000:11, 16).

Figure 9.

Levanna points from 36Mr133 that tested positive for protein residues. Left: FS#131. Right: FS#185. Modified from Hornum et al. (2009: Figure 47).

In summarizing research dealing with Late Woodland triangular projectile points, Engelbrecht (2014) notes that the lack of notches implicates the probable use of adhesives rather than bindings in securing the bifaces to a shaft. This approach provides advantages for the use of triangular bifaces in a bow and arrow system. The lack of binding promotes penetration of the projectile. The points easily detach from the shaft making it possible to retrieve shafts which are more costly to produce than the bifaces. Given these assumptions about hafting, triangular bifaces may have been too inefficient as knives and were always intentionally produced as projectiles. This does not negate the fact that some may have been used on an expedient basis as cutting implements while secured in a shaft.

The dramatic decrease in the variability of biface forms/types during the Late Woodland period has been linked with the widespread adoption of bow and arrow technology (Custer, 2001:113). What remains problematic is the occurrence of nearly identical triangular forms much earlier in time when a variety of biface types characterize assemblages in the region. If we accept that earlier triangles reflect the presence of the bow, than we must assume that the technology was not viewed as sufficiently superior to other types of projectile technologies to result in their replacement. Perhaps a more reasonable interpretation would be to view early triangles as one of many different types of atlatl dart tips, perhaps for use with specific types of prey. A large-scale program of residue analysis is necessary to address this possibility. Additional discussion of this topic is presented below.

Conclusion

Obviously radiocarbon dates cannot be considered in a vacuum and some of the associations tabulated in this report could be challenged as problematic. However, we must not blind ourselves to the potential contemporaneous use of a number of biface types (discussion in Stewart, 2018). Explaining typological diversity as a result of functional differences may best be resolved by extensive residue analysis to see if a specific form can be related to specific prey or the harvesting and processing of other identifiable resources. To-date, residue and microwear analyses most frequently imply that biface types are multipurpose, but sample sizes and the range of types represented in these studies are small. Perhaps some biface forms are gender-specific as certain types of labor and lithic technology may be (e.g., Sassaman, 1992; Waguespack, 2005). Some portion of biface variability for the Late Archaic through Middle Woodland periods likely relates to heightened social interaction. We should closely consider changes through time in the means of travel (foot, boat) and the travel corridors employed (Delaware River and major tributaries, Ridge and Valley system) and the impact that they might have had on the extent and patterns of interaction. Is the diversity of biface types in contemporaneous assemblages greater on sites located within natural transportation corridors, especially sites representing prolonged habitation where local and travelling groups might be expected to interact? Does typological diversity exhibit a falloff pattern with increasing distance from such settings? Simply invoking trade to account for some assemblage variation falls short of what we might otherwise learn.

We presume that reported typological assessments adhere to a rigorous comparison to original type definitions, rather than an investigator’s personal view of the variation that could be embraced by any given type. The latter is more likely the case; variability is a given in the use history of a biface regardless of the mental template in force at the time of its creation. The question remains, at what point does the variability that we observe actually relate to a single, and chronologically distinct template? Given the variability and overlap of triangular points of Archaic origin from the Delaware Valley and broader region, it does not seem useful to continue to use existing type names (i.e., Beekman, Squibnocket, and Hunter Brook) to characterize future finds. Continuing to compare and contrast the attributes of new finds with standard typological descriptions is useful, but such endeavors should also reflect on the additional data and comparative samples discussed here.

Regardless of the potential problems with some of the data reported here, it is clear that relatively small, triangular bifaces are in use long before 600/900 AD and the presumed advent of the bow and arrow in the region. The morphological and functional similarity of Archaic and Late Woodland triangles raises the issue of whether bow and arrow technology existed in the region earlier than typically presumed since Late Woodland triangles are typically considered to be arrowheads (Custer, 2001: 87; Katz, 2000: 101). In studies that have used ethnographic examples of arrowheads as a baseline for determining relevant attributes of projectile points, weight and width are important in proposed discriminant functions (Katz, 2000: 101 citing Shott, 1997 and Thomas, 1978). The attributes of Archaic and Late Woodland triangles from the Delaware Valley exhibit an overlap with those associated with arrow heads.

Reductions in projectile size, shoulder or basal width, and thickness remain key variables in the indirect argument for the presence of bow and arrow technology (Blitz and Porth, 2013: 89–91). Projectiles with greater mass, such as those that might be used on a spear, provide the greatest penetration and wounding potential while the light weight and greater velocity typical of the bow and arrow produces a flatter trajectory and better accuracy (Yu, 2006: 207). Custer (1991: 58–63; 2001: 87) provides a comparative discussion of the “stopping power” or kinetic energy of spear and spearthrower versus bow and arrow, noting that kinetic energy is a product of the projectile’s weight and the square of its velocity. Increasing a projectile’s velocity while holding its mass constant, something made possible by the use of the bow, dramatically magnifies the kinetic energy of a projectile. “Because of the higher velocity of arrows, a smaller point could be used; and a smaller, lighter projectile would increase velocity and resultant kinetic energy” (Custer, 2001: 87). While wounding and penetration are important, the ability to accurately target a specific area of an animal’s body may be more so (Custer, 1991: 58) and is a benefit of bow and arrow technology.

Debate concerning the origins of bow and arrow technology in North America is ongoing, including the possibility that it appears millennia before 300 – 600 AD, the time typically associated with the technology in the Eastern Woodlands (Blitz and Porth, 2013; see summary in Toner, 2015). In the Delaware Valley a Middle Woodland origin, ca. 500/600 AD, is presumed as represented by Jacks Reef points (Kraft, 2001: 194). The use of the bow and arrow, regardless of its antiquity, does not preclude the continued use of the demonstrably more ancient spearthrower or atlatl, as is seen ethnographically in the Southwestern United States and Central America (Toner, 2015; Yu, 2006: 201).

Exploring the possible adoption of the bow and arrow prior to 500/600 AD in the Delaware Valley and broader region is important given the attendant implications. Individual rather than cooperative hunting may have been favored, impacting the sharing of prey and the social contexts involved (Grund, 2017; Toner, 2015). To the degree that the bow and arrow contributed to greater household autonomy the size and structure of residential camps may have changed, as well as the nature of archaeological sites construed as hunting stations or hunting camps (Blitz and Porth, 2013; Toner, 2015). The potential for increased violence or coercion and related social impacts has been linked with the adoption of the bow (Blitz and Porth, 2013). Future research also needs to consider that pointed fragments of debitage as well as bone and antler can be effective arrow points (e.g., Knecht, 1997; Odell, 1988; Odell and Cowan, 1986; Waguespack et al., 2009). Detailed use wear and comparative studies of triangles of all ages will be crucial in the future exploration of this idea.

Attributes that Katz (2000: 80–81) indicates distinguish Archaic from Late Woodland triangles are not a matter of presence or absence, but their degree of representation in a sample. Studies of biface morphology and the technological recipes used in their production may yet define ways to isolate early from late triangles. Working out the difficulties of sorting types with lengthy chronological ranges could be pursued from the standpoint of technological style (or chaîne opératoire) which examines the details of the entire manufacturing process from the selection of raw material, core or blank production, and patterned sequences of flake removals. Models for how to proceed with such work are known for the region (e.g., Bergman et al., 1996b: 207–210). Such studies must recognize that approaches to production could vary by geographic area, reflecting specific learning networks of knappers. Thus, a suite of useful distinguishing characteristics in one area might not be effective in another.

Settlement pattern studies, so reliant on surface and plowzone sites, must consider the impact of the extensive use history of triangular bifaces. The same holds for assigning age to contexts solely on the basis of the biface types recovered. Hopefully, the current study provides a basis on which future interpretations of assemblages and other research endeavors can build.

Footnotes

Acknowledgments

Major funding for the research reported here was provided by the New Jersey Historic Preservation Office of the state’s Department of Environmental Protection. Many thanks to Kate Marcopul, State Historic Preservation Officer and Administrator, who initially envisioned the overall project of which this article is a part and made it a reality. Thanks also to Vincent Maresca, Senior Historic Preservation Specialist at the New Jersey Historic Preservation Office, for providing information that I missed during my initial review of cultural resource management (CRM) reports. Earlier related projects were supported by the Federal Highway Commission and the New Jersey Department of Transportation. An extensive list of colleagues provided assistance in numerous ways to the overall project from which this article derives. They include: Keith Bastianini, Del Beck, Kurt Carr, Peter N. Chletsos, Mike Clem, Jay Custer, Tommy Davies, Andy Dillman, Joseph A.M. Gingerich, Adam Heinrich, Richard Hunter, Don Kline, Jeremy Koch, Gregory Lattanzi, Jonathan Lothrop, Doug McLearen, Roger Moeller, Alison Oskam, Mike Owens, Walt Padpora, John Parks, Philip Perazio, George Pevarnik, Lori Rohrer, Peter Siegel, Mark Shaffer, Noel Strattan, Edwin Struve, Matt Tomaso, Don Troxell, Rob Tucher, Rich Veit, Nina Versaggi, Elizabeth Wagner, and Leonard Ziegler. I am grateful to all for their help.

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) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

R Michael Stewart

Author Biography

R Michael Stewart has a long record of research and publication dealing with the archaeology of American Indians ranging from the Paleoindian period (13,000 years ago) through native contact and interaction with Europeans in the Mid-Atlantic and Northeast regions. His research interests include human ecology, trade, pottery technology, and the integration of archaeology and environmental sciences. Stewart has degrees in Anthropology from the University of Delaware (BA) and the Catholic University of America (MA, PhD). He is emeritus faculty of the Department of Anthropology at Temple University in Philadelphia and currently works part-time as an archaeologist for the New Jersey Historic Preservation Office. He provides consulting services through Ancient Lives Investigations. His latest work focused on the Upper Delaware Valley (New Jersey, Pennsylvania, and New York) can be accessed on the website of the New Jersey Historic Preservation Office at

References

Anderson

Bastianini

Filkins

, et al. (2000) Treichlers Bridge (36-Nm-142), Phase III Archaeological Data Recovery, Northampton County, Pennsylvania, Volumes I, IIA, IIB and IV. Report on file (ER#88-0481-042), Pennsylvania State Historic Preservation Office, Harrisburg.

Bello

Stewart

Digiugno

GUY

(1998) Archaic-period triangles from 28GL228. Bulletin of the Archaeological Society of New Jersey 53: 128.

Bergman

Doershuk

Kimball

, et al. (1996a) Archaeological Data Recovery for Transcontinental Gas Pipe Line Corporation’s 6.79 Mile Leidy Natural Gas Pipeline Expansion, Sandts Eddy Site (36-NM-12), Northampton County, Pennsylvania. Volumes I-IV. Report on file (ER#92-0705-095), Pennsylvania State Historic Preservation Office, Harrisburg.

Bergman

Russell

Duerksen

, et al. (1996b) Analysis of the prehistoric lithic assemblages. In: Bergman CA, Doershuk J, Kimball LR, et al. (eds) Archaeological Data Recovery for Transcontinental Gas Pipe Line Corporation’s 6.79 Mile Leidy Natural Gas Pipeline Expansion, Sandts Eddy Site (36-NM-12), Northampton County, Pennsylvania, Volume II, pp. 90–210. Report on file (ER#92-0705-095), Pennsylvania State Historic Preservation Office, Harrisburg.

Bergman

Carr

Adovasio

(2020) The early and middle archaic periods of Pennsylvania. In: Carr

Bergman

Rieth

, et al. (eds) The Archaeology of Native Americans in Pennsylvania. Philadelphia: University of Pennsylvania Press, pp. 105–139.

Blitz

Porth

(2013) Social complexity and the bow in the Eastern Woodlands. Evolutionary Anthropology 22: 89–95.

Cavallo

(1981) Turkey swamp: A Late Paleo-Indian Site in New Jersey’s coastal plain. Archaeology of Eastern North America 9: 1–17.

Church

(2000) Appendix IV: Results of a high power use-wear analysis of a portion of the lithic assemblage from the Treichlers Bridge Site (36NM142), Northampton County, Pennsylvania. In: Anderson DA, Bastianini KR, Filkins E, et al. (eds) Treichlers Bridge (36-Nm-142), Phase III Archaeological Data Recovery, Northampton County, Pennsylvania, Volume III. Report on file (ER#88-0481-042), Pennsylvania State Historic Preservation Office, Harrisburg.

Cremeens

Hart

(2009) Holocene alluvial geomorphology of the Memorial Park Site (36Cn164), West Branch Susquehanna River, Pennsylvania. Archaeology of Eastern North America 37: 47–64.

10.

Custer

(1983) Quantitative Analysis of the Shenks Ferry and Susquehannock Projectile Points: Testing the Inter-Ethnic Variability of Stone Tools (Report No. 1). Newark: University of Delaware Center for Archaeological Research.

11.

Custer

(1991) Notes on broadspear functions. Archaeology of Eastern North America 19: 51–73.

12.

Custer

(1996) PrehistoricCultures of Eastern Pennsylvania. Pennsylvania Historical and Museum Commission, Anthropological Series No. 7, Harrisburg.

13.

Custer

(2001) ClassificationGuide for Arrowheads and Spearpoints of Eastern Pennsylvania and the Central Middle Atlantic. Harrisburg: Pennsylvania Historical and Museum Commission.

14.

Custer

Rosenberg

Mellin

, et al. (1990) A re-examination of the Island Field Site (7K-F-17). Archaeology of Eastern North America 18: 145–212.

15.

Custer

Watson

BAILEY

(1994) Data Recovery Investigations of the West Water Street Site, 36CN175, Lock Haven, Clinton County, Pennsylvania. Report on file, Pennsylvania State Historic Preservation Office, Harrisburg.

16.

Custer

Watson

Bailey

(1996) A summary of phase III data recovery excavations at the West Water Street Site (36CN175), Lock Haven, Clinton County, Pennsylvania. Pennsylvania Archaeologist 66(1): 1–53.

17.

Delaware River Basin Commission (2018) Watersheds of the Delaware River Basin. West Trenton: Delaware River Basin Commission

18.

Doershuk

(1994) Spatial analysis of artifacts and features. In: C. Bergman et al. (eds) Archaeological Data Recovery for Transcontinental Gas Pipe Line Corporation's 6.79 Mile Leidy Natural Gas Pipeline Expansion, Padula Site (36-NM-15), Northampton County, Pennsylvania, pp. 299–357. Report on file (ER# 82-0648-042), Pennsylvania State Historic Preservation Office, Harrisburg.

19.

Doucette

(2005) Reflections of the middle archaic: A view from annasnappet pond. Bulletin of the Massachusetts Archaeological Society 66(1): 22–33.

20.

East

Langer

Espenshade

, et al. (2002) Northumberland and Union Counties, Pennsylvania, I80, Section 52D, Bridge Expansion and Highway Improvement Project, Phase I/II/III Archaeological Investigations, E.R. #99-8000-042. Report on file, Pennsylvania State Historic Preservation Office, Harrisburg.

21.

Engelbrecht

(2014) Unnotched triangular points on village sites. American Antiquity 79(2): 353–367.

22.

Engelbrecht

(2015) Interpreting broken arrowpoints. American Antiquity 80(4): 760–766.

23.

Funk

(1965) The archaic of the Hudson Valley – new evidence and interpretations. Pennsylvania Archaeologist 35(3-4): 139–160.

24.

Funk

(1975) An evaluation from the New York perspective. In: Kraft

(ed) The Archaeology of the Tocks Island Area. South Orange: Archaeological Research Center, Seton Hall University Museum, pp. xiii–xv.

25.

Funk

(1976) Recent Contributions to Hudson Valley Prehistory. New York State Museum Memoir 22, Albany.

26.

Funk

(1988) The Laurentian concept: A review. Archaeology of Eastern North America 16: 1–42.

27.

Funk

(1989) Some contributions of archaeology to the study of cave and Rockshelter sediments: Examples from Eastern New York. Man in the Northeast 37: 35–112.

28.

Funk

(1993) Archaeological Investigations in the Upper Susquehanna Valley, New York State, Volume 1. Buffalo: Persimmon Press.

29.

Funk

(1998) Archaeological Investigations in the Upper Susquehanna Valley, New York State, Volume 2. Buffalo: Persimmon Press.

30.

Funk

Walters

Scott

(1971) A stratified Rockshelter in the Upper Delaware Valley. Man in the Northeast 2: 23–50.

31.

Funk

Wellman

(1984) Evidence of early Holocene occupation in the Upper Susquehanna Valley, New York State. Archaeology of Eastern North America 12: 81–109.

32.

Goetz

Jantz

Sun

(2011) Forecasting future land use and its hydrologic implications: A case study of the upper Delaware river watershed. Watershed Science Bulletin Fall: 18–26.

33.

Grund

(2017) Behavioral ecology, technology, and the organization of labor: How a shift from spear thrower to self bow exacerbates social disparities. American Anthropologist 119(100): 104–119.

34.

Hartgen Archaeological Associates, Inc. (1988a) Peake Historic and Prehistoric Site, 12-68-7. Macy South 345 kV Transmission Line, Town of Delhi, Delaware County, N.Y. Report on file, New York State Office of Parks, Recreation and Historic Preservation, Cultural Resource Information System, Albany.

35.

Hartgen Archaeological Associates, Inc. (1988b) Otego Yard Historic and Prehistoric Site, NYSM 121. Marcy South 345 kV Transmission Line, Town of Otego, Otsego County, New York (2 volumes). Report on file, New York State Office of Parks, Recreation and Historic Preservation, Cultural Resource Information System, Albany.

36.

Hartgen Archaeological Associates, Inc. (1989) Ouleout Site, 8W-32-7. Marcy South 345 kV Transmission Line, Town of Franklin, Delaware County, New York. Report on file, New York State Office of Parks, Recreation and Historic Preservation, Cultural Resource Information System, Albany.

37.

Heinrich

(2018) Phase III Archaeological Data Recovery. The Sorbello 1 Site (28-SA-214), Straughns Mill Road, Block 28.01, Lot 52, Oldmans Township, Salem County, New Jersey. Report on file, New Jersey Historic Preservation Office, Trenton.

38.

Heinrich

Hinshaw

(2017) Phase II Archaeological Survey of the Sorbello North Site (28-Sa-213) and the Sorbello 1 Site (28-Sa-214), Straughns Mill Road, Block 28.01, Lot 52 Oldmans Township, Salem County, New Jersey. Report on file, New Jersey Historic Preservation Office, Trenton.

39.

Hennessy

(1992) 1990 Archaeological Excavations at the Pardee Site, 36-MR-5. Smithfield Beach Development Area, Delaware Water Gap National Recreation Area, Pennsylvania. Report on file, National Park Service, Mid-Atlantic Region, Philadelphia, Pennsylvania.

40.

Hohman

Versaggi

(2003) Phase II Site Examinations, Herrick Hollow II, Herrick Hollow III, Herrick Hollow IV, Herrick Hollow V, Herrick Hollow VI, Herrick Hollow VII. Richardson Hill Road Landfill Superfund Project, Town of Masonville, Delaware County, New York. Report on file (Digital Archive ID# 3298), New York State Office of Parks, Recreation and Historic Preservation, Cultural Resource Information System.

41.

Hornum

Laporta

ROTH

, et al. (2009) Phase III Archaeological Data Recoveries at 36NM204, 36MR119, And 36MR133 for the Proposed Line 1278 Replacement Project, Northampton and Monroe Counties, Pennsylvania. ER# 97-2182-042-PP. Three volumes. Report on file, Pennsylvania State Historic Preservation Office, Harrisburg.

42.

Inashima

(2008) Establishing a radiocarbon date based framework for Northeastern Virginia archaeology. Quarterly Bulletin of the Archeological Society of Virginia 63(4): 187–290.

43.

Inashima

(2011) Supplement no. 1 to “establishing a radiocarbon date based framework for Northeastern Virginia archaeology.” Quarterly Bulletin of the Archeological Society of Virginia 66(3):89–140.

44.

Johnson

Mahlstedt

Bradley

, et al. (1984) Guide to Prehistoric Site Files and Artifact Classification System. Boston: Massachusetts Historical Commission, Office of the Massachusetts Secretary of State.

45.

Johnson

(2002) Prehistoric context. In: Fetzer R, Fuess T, Johnson, et al. (eds) Phase I Archaeological and Intensive Architectural Survey of Route 94 Between Mount Pleasant Road, Knowlton Township, Warren County, New Jersey, 2-13 to 2-71. Report on file (WAR F591-6049), New Jersey Historic Preservation Office, Trenton.

46.

Justice

(1995) Stone Age Spear and Arrow Points of the Midcontinental and Eastern United States. Bloomington: Indiana University Press.

47.

Katz

(2000) Archaic period triangular bifaces in the Middle Atlantic Region: Technological and functional considerations. MA Thesis, Department of Anthropology, Temple University, USA.

48.

Kimball

(1996) Microwear analysis of the functions of projectile points and flake tools. In: Bergman CA, Doershuk J, Kimball LR, et al. (eds) Archaeological Data Recovery for Transcontinental Gas Pipe Line Corporation’s 6.79 Mile Leidy Natural Gas Pipeline Expansion, Sandts Eddy Site (36-NM-12), Northampton County, Pennsylvania, Volume II, pp. 211–291. Report on file (ER#92-0705-095), Pennsylvania State Historic Preservation Office, Harrisburg.

49.

Kinsey

III (1972) Archeology in the Upper Delaware Valley. Pennsylvania Historical and Museum Commission, Anthropological Series, No.3, Harrisburg.

50.

Kinsey

III (1975) Faucett and Byram Sites: Chronology and Settlement in the Delaware Valley. Pennsylvania Archaeologist 45(1–2): 1–103.

51.

Knapp

(2011) Life at the Confluence: Archaic and Woodland Occupations of the Chenango Point Site (SUBi-1274). Data Recovery Excavations for the Downtown Academic Center Project, City of Binghamton, Broome County, New York. Report on file (Digital Archive), New York State Office of Parks, Recreation and Historic Preservation, Cultural Resource Information System.

52.

Knecht

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

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

53.

Kotcho

(1998) A discriminant analysis of late archaic/early woodland and middle/late woodland period projectile points. Paper prepared for the class, Research Methods and Theory in Archaeology: Lithic Analysis. Department of Anthropology, Rutgers University, New Brunswick, New Jersey. On file, New Jersey State Museum, Department of Archaeology and Ethnology, Trenton.

54.

Kraft

(1975) The Archaeology of the Tocks Island Area. South Orange: Archaeological Research Center, Seton Hall University.

55.

Kraft

(1976) The Rosenkrans Site, an adena-related mortuary complex in the Upper Delaware Valley, New Jersey. Archaeology of Eastern North America 3: 9–50.

56.

Kraft

(2001) The Lenape-Delaware Indian Heritage: 10,000 BC to AD 2000. Stanhope: Lenape Books.

57.

Kudrle

(2005) Sidney Municipal Airport Project: Phase 3 Data Recovery, Sidney Hangar Site (SUBi-2073), Town of Sidney, Chenango and Delaware Counties, New York. Report on file (Digital Archive), New York State Office of Parks, Recreation and Historic Preservation, Cultural Resource Information System.

58.

Lee

Martin

Stiteler

(2010) Phase III Archaeological Data Recovery, Site 28WA290, Route I-80 Truck Weigh Station Project, Knowlton Township, Warren County, New Jersey. Report on file (WAR 704a-9789), New Jersey Historic Preservation Office, Trenton.

59.

Leslie

(1963) A typology of arrowpoints found in Wayne County, Pennsylvania, and in Sullivan County, New York. New World Antiquity 10(7–8): 70–90.

60.

Lothrop

Koldehoff

(1994) Archaic settlement in the Lower Delaware Valley. Journal of Middle Atlantic Archaeology 10: 105–123.

61.

Louis Berger and Associates, Inc. (1992) SS-2 and APEC Storage Service Projects. Woodbury Lateral Loop Natural Gas Pipeline Expansion. Archaeological Investigations at Sites 28GL111, 28GL205, 28GL206, 28GL209 and 28GL210. West Deptford and East Greenwich Townships, Gloucester County, New Jersey. Report on file, New Jersey Office of Historic Preservation, Trenton.

62.

Lowery

(2013) Jack’s Reef in the Chesapeake and Delmarva Region: Research into the Coastal Archaeology of the Era Between circa cal A.D. 480 and cal A.D. 900. Archaeology of Eastern North America 40: 3–30.

63.

Luckenbach

Grow

Sharpe

(2010) Archaic period triangular points from pig point, Anne Arundel County, Maryland. Journal of Middle Atlantic Archaeology 26: 165–179.

64.

Marchiando

(1972) Bell-browning site, 28-Sx-19. In: Kinsey

III (ed) Archeology in the Upper Delaware Valley ( Anthropological Series Number 2). Harrisburg: Pennsylvania Historical and Museum Commission, pp. 131–158.

65.

Mcconaughy

(2013) The Jack’s reef horizon in Pennsylvania: A preliminary assessment. Archaeology of Eastern North America 40: 31–46.

66.

Mclearen

Fokken

(1986) White Horse West Site (28ME119), Data Recovery. Trenton Complex Archaeology Report 4. Report on file, New Jersey Department of Transportation and the New Jersey Office of Historic Preservation, Trenton.

67.

Miller

(1994) Archaeological Data Recovery, The Skvarek Site (Site 36Lu132), Proposed Shickshinny/Mocanaqua Bridge Replacement (S.R. 0239), Salem and Conyngham Townships, Luzerne County, Pennsylvania. Report on file, Pennsylvania State Historic Preservation Office, Harrisburg.

68.

Miller

(1998) Lithic projectile point technology and raw material use in the Susquehanna River Valley. In: Raber PA, Miller PE and Neusius SM (eds) The Archaic Period in Pennsylvania: Hunter-Gathers of the Early and Middle Holocene Period (Recent Research in Pennsylvania Archaeology Number 1). Harrisburg: Pennsylvania Historical and Museum Commission, pp. 91–120.

69.

Miller

Marine

Woodard Mcknight

, et al. (2010) Archaeological investigations on Calver Island (36Da89): A stratified site in the Lower Susquehanna River Valley. Journal of Middle Atlantic Archaeology 26: 81–103.

70.

Miller

Vento

Marine

, et al. (2009) Phase III Archaeological Data Recovery, Route 11/15 Improvements (SR 0011, Section 008), Juniata and Perry Counties, Pennsylvania, Volumes 1-6. Report on file, Pennsylvania State Historic Preservation Office, Harrisburg.

71.

Miroff

(2012) Investigation of the Chenango Point South Site (SUBi-2776). Phase 2/3 Archaeological Investigations, Chenango Point South Site, Twin River Commons Project, City of Binghamton, Broome County, New York. Report on file (Digital Archive), New York State Office of Parks, Recreation and Historic Preservation, Cultural Resource Information System.

72.

Miroff

(2014) The Otsiningo Market Site (SUBi-3041; A00706.000048), Phase 3 Archaeological Investigations, Otsiningo Farmers Market Project, Town of Dickinson, Broome County, New York. Report on file (Digital Archive), New York State Office of Parks, Recreation and Historic Preservation, Cultural Resource Information System.

73.

Mounier

(2003) Looking Beneath the Surface: The Story of Archaeology in New Jersey. New Brunswick: Rutgers University Press.

74.

Mounier

(2008) The Aboriginal Exploitation of Cuesta Quartzite in Southern New Jersey. PhD Dissertation, Department of Anthropology and Archaeology, Memorial University of Newfoundland, St. John’s Newfoundland.

75.

Odell

(1988) Addressing prehistoric hunting practices through stone tool analysis. American Anthropologist 90: 335–355.

76.

Odell

Cowan

(1986) Experiments with spears and arrows on animal targets. Journal of Field Archaeology 13(2): 195–212.

77.

Parr

(2006) Protein residue analysis of artifacts from archaeological sites 36MR133, 36MR119, and 36NM204, Monroe and Northampton Counties, Pennsylvania. Appendix 7. In: Hornum MB, Laporta PC, Roth J, et al. (eds) Phase III Archaeological Data Recoveries at 36NM204, 36MR119, And 36MR133 for the Proposed Line 1278 Replacement Project, Northampton and Monroe Counties, Pennsylvania. ER# 97-2182-042-PP. Three volumes. Report on file, Pennsylvania State Historic Preservation Office, Harrisburg.

78.

Pennsylvania Fish and Boat Commission (2011) Delaware River Management Plan: A Management Plan Focusing on the Large River Habitats of the West Branch Delaware River and Non-tidal Reach of the Delaware River of Pennsylvania. Harrisburg: Bureau of Fisheries, Division of Fisheries Management.

79.

Puniello

(1980) Iroquois series ceramics in the Upper Delaware Valley. In: CF

Hayes

III (ed) Proceedings of the 1979 Iroquois Pottery Conference ( Rochester Museum and Science Service, Research Record No. 13). New York: Rochester, pp. 146–155.

80.

Reimer

Bard

Bayliss

, et al. (2013) INTCAL13 and marine13 radiocarbon age calibration curves 0–50,000 years cal BP. Radiocarbon 55(4): 1869–1887.

81.

Reinbold

Lawrence

Bailey

(2000) Phase III Data Recovery 36LY290: S.R. 0405, Section 61M, Wolf Township, Lycoming County, Pennsylvania. Report on file, Pennsylvania State Historic Preservation Office, Harrisburg.

82.

Rieth

(2013) Space, time and the middle woodland “Jack’s reef horizon” in New York. Archaeology of Eastern North America 40: 91–112.

83.

Ritchie

(1961) A Typology and Nomenclature for New York Projectile Points (Bulletin Number 384). Albany: New York State Museum and Science Service.

84.

Ritchie

(1965) The Archaeology of New York State. Garden City: Natural History Press.

85.

Ritchie

(1971) A Typology and Nomenclature for New York Projectile Points (Bulletin Number 384). Albany: New York State Museum and Science Service (revision of 1961 edition).

86.

Ritchie

Funk

(1973) Aboriginal Settlement Patterns in the Northeast (Memoir 20). Albany: New York State Museum and Science Service.

87.

Sassaman

(1992) Lithic technology and the hunter-gatherer sexual division of labor. North American Archaeologist 13(3): 249–262.

88.

Shott

(1997) Stones and shafts redux: The metric discrimination of chipped-stone dart and arrow points. American Antiquity 62(1): 86–101.

89.

Siegel

Benedict

Kingsley

(1999) Archeological Data Recoveries at the Fahs II and Oberly Island Sites: Structure, Function, and Context in the Lower Lehigh Valley Northampton County, Pennsylvania S.R. 0033, Section 001. Report on file (ER# 88-0224-095), Pennsylvania State Historic Preservation Office, Harrisburg.

90.

Siegel

Kingsley

Benedict

(2001) Oberly island: Trend and tradition in the Lower Lehigh Valley. Pennsylvania Archaeologist 71(2): 16–59.

91.

Snow

(1980) The Archaeology of New England. New York: Academic Press.

92.

Snyder

Petyk

(2010) Results of Cultural Resource Survey for the Proposed 300 Line Project in Pennsylvania. Draft Addendum for Phase II Evaluations of Sites 36WY0159 and 36PI0239. Report on file (ER# 08-2473-042), Pennsylvania State Historic Preservation Office, Harrisburg.

93.

Stewart

(1987) Gropp's Lake Site (28Me100G), Data Recovery. Trenton Complex Archaeology Report 2. Report on file, New Jersey Department of Transportation and New Jersey Historic Preservation Office, Trenton (Revised edition 1996).

94.

Stewart

(1990a) Lost Cultures Found: New Evidence for Middle Archaic Occupations in the Delaware Valley. Paper presented at the annual meeting of the Society for Pennsylvania Archaeology, Wilkes Barre.

95.

Stewart

(1990b) New Evidence of the Middle Archaic Period in the Delaware Valley. Paper presented at the annual meeting of the New York State Archaeological Association, Sparrow Bush, New York.

96.

Stewart

(1998) Archaic Triangles at the Abbott Farm National Landmark: Typological Implications for Prehistoric Studies in the Middle Atlantic Region. Paper accompanying an exhibit of Archaic-Age Bifaces at the annual meeting of the Middle Atlantic Archaeological Conference, Cape May, New Jersey. On file, New Jersey State Museum, Department of Archaeology and Ethnology, Trenton.

97.

Stewart

(2018) A Radiocarbon Foundation for Archaeological Research in the Upper Delaware Valley: New Jersey, Pennsylvania, New York. Report prepared for the New Jersey Historic Preservation Office, Trenton.

98.

Stewart

Adovasio

Shaunessy

, et al. (2000) Phase III Data Recovery, Howarth-Nelson Site (36FA40), Fayette County, Pennsylvania. Report on file, Pennsylvania State Historic Preservation Office, Harrisburg.

99.

Stewart

Cavallo

(1991) Delaware valley middle archaic. Journal of Middle Atlantic Archaeology 7: 19–42.

100.

Stewart

Messner

Perazio

, et al. (2015) The Archaeology of the Manna Site (36Pi4), Delaware Water Gap National Recreation Area, Pike County, Pennsylvania. Report on file, National Park Service, Delaware Water Gap National Recreation Area, Bushkill, Pennsylvania.

101.

Stuiver

Reimer

(1993) Extended (super 14) C data base and revised CALIB 3.0 (super 14) C age calibration program. Radiocarbon 35(1): 201–213.

102.

Stuiver

Reimer

(2017) CALIB 7.1 (WWW program). Available at: (http://calib.org accessed 17 August 2020).

103.

Thomas

(1978) Arrowheads and atlatl darts: How the stones got the shaft. American Antiquity 43(3): 461–480.

104.

Toner

(2015) From atlatls to arrows. American Archaeology 19(1): 19–25.

105.

USGS National Water Information System (2018) USGS Water Resources Links for: 02040101 – Upper Delaware . Available at: https://water.usgs.gov/lookup/getwatershed?02040101 (accessed 17 August 2020).

106.

Waguespack

(2005) The organization of male and female labor in foraging societies: Implications for early Paleoindian archaeology. American Anthropologist 107(4): 666–676.

107.

Waguespack

Surovell

Denoyer

, et al. (2009) Making a point: Wood-versus stone-tipped projectiles. Antiquity 83(321): 786–800.

108.

Walker

(2013) An examination of Jack’s reef in New Jersey. Archaeology of Eastern North America 40: 47–58.

109.

Wall

(2015) The end of the late archaic period in the Upper Susquehanna Valley of Pennsylvania. In: Stewart RM, Carr K and Raber P (eds) The Nature and Pace of Change in American Indian Cultures: Pennsylvania, 3000 to 4000 BP (Recent Research in Pennsylvania Archaeology, Volume 4). University Park: Penn State Press, pp. 37–52.

110.

Wall

Botwick

(1995) Archaeological Survey of the Milford Transect, Delaware Water Gap National Recreation Area, Pennsylvania and New Jersey. Volume I: Report Text. Report on file, National Park Service, Delaware Water Gap National Recreation Area, Bushkill, Pennsylvania.

111.

Wall

Stewart

Cavallo

(1996a) The Lithic Technology of the Trenton Complex. Trenton Complex Archaeology Report 13. Report on file, New Jersey Department of Transportation and New Jersey Historic Preservation Office, Trenton.

112.

Wall

Stewart

Cavallo

, et al. (1996b) Area D Site (28Me1-D), Data Recovery. Trenton Complex Archaeology Report 9. Report on file, New Jersey Department of Transportation and New Jersey Historic Preservation Office, Trenton.

113.

Williams

Puniello

Flinn

(1982) Reinvestigation of the Late Woodland Occupation in the Delaware Water Gap National Recreation Area, New Jersey. Report on file (E78), New Jersey Historic Preservation Office, Trenton.

114.

Wingerson

(1976) The Hunter Brook Rockshelter. Bulletin of the New York State Archaeological Association 68: 19–28.

115.

(2006) From atlatl to bow and arrow: Implicating projectile technology in changing systems of hunter-gather mobility. In: Sellet

Greaves

(eds) Archaeology and Ethnoarchaeology of Mobility. Gainesville: University Press of Florida, pp. 201–220.