midcontinental journal of archaeology, 2016, 1–34
Petrographic Analysis of Late Woodland
and Middle Mississippian Ceramics at
the Iva Site (47Lc42), Onalaska,
Wisconsin
Robert F. Boszhardt and James B. Stoltman
University of Wisconsin-Madison, USA
The Iva site contained a rare effigy mound and Middle Mississippian (Ramey
horizon) component within the Late Woodland Lewis phase territory of the
Upper Mississippi River valley. Salvage excavations in 2002–2003 recovered
fragments of numerous Angelo Punctated, Powell Plain, and Ramey Incised
vessels, including examples of Angelo and Ramey in direct association. Petrographic analysis was conducted on seven grit-tempered and six shell-tempered
vessels, eight of which are stylistically Mississippian. The results indicate that
four of eight Mississippian vessels were likely manufactured in the American
Bottom, with the other half being local imitations of Mississippian styles.
These data are compared to contemporaneous Ramey horizon components
in the Driftless Area of Cahokia’s northern hinterland.
keywords Mississippian, effigy mounds, interaction, Angelo Punctated,
Ramey horizon
Introduction
In the centuries preceding the onset of the Middle Mississippian culture in ca. A.D.
1050, much of the upper Midwest was occupied by relatively insular Late Woodland
societies (Emerson et al. 2000). Late Woodland peoples of southern and western Wisconsin were distinct in that they constructed effigy mounds and manufactured local
ceramic wares. Differential distributions of certain effigy mound forms and associated
diagnostics indicate various subgroups (Boszhardt and Goetz 2000; Rosebrough
2010). These are sometimes separated by discernible territorial boundaries that seem
to reflect population increases and socioeconomic stress (Theler and Boszhardt 2006).
The rather sudden appearance of “Middle Mississippian” was centered on the
ancient metropolis of Cahokia near modern St. Louis (Pauketat 2009). Coinciding
© Midwest Archaeological Conference Inc 2016
DOI 10.1080/01461109.2015.1114251
2
ROBERT F. BOSZHARDT AND JAMES B. STOLTMAN
with the transition from the Edelhardt to the Lohmann phase in the American
Bottom (ca. A.D. 1050), Mississippians ventured nearly 875 km up the Mississippi
River to the Fisher Mounds Site Complex at Stoddard, Wisconsin (Benden 2004;
Stoltman et al. 2008). At about the same time, a more substantial Mississippian
colony was established 45 km farther upriver at Trempealeau, Wisconsin (Benden
et al. 2011; Boszhardt and Benden 2014; Boszhardt et al. 2012, 2013, 2015;
Green and Rodell 1994; Pauketat et al. 2015). After A.D. 1100, and corresponding
with the early portion of the Stirling phase at Cahokia (A.D. 1100–1150), a second
wave of Mississippian influences spread north from Cahokia to Wisconsin, northern
Illinois, Iowa, and southern Minnesota in what is referred to as the Ramey horizon
(Kelly 1991a; Pauketat 2004:11–12).
Why did Mississippians venture northward and how did they interact with indigenous Late Woodland populations? These questions have been the basis of interaction model development and testing in the northern hinterland for well over
half a century. (See Green [2015] and Millhouse [2012] for recent historical overviews of this topic.) Such models have been revised as new sites were located and
assemblages described, chronologies refined both in the American Bottom and at
related sites to the north, and analytical technologies advanced to allow clarification
of whether diagnostic artifacts were imported or locally made. One such technique is
petrographic analysis of thin sections from ceramic vessels. Salvage excavations at
the Iva site in western Wisconsin produced a mixed Late Woodland and Middle Mississippian assemblage (Boszhardt 2004). Petrographic analysis of a sample of 13 Iva
vessels is reported here. The results provide insight into when and how interaction
occurred at this site and how this fits into Mississippian interaction models for
the Upper Mississippi River valley (UMRV).
Background
In the unglaciated Driftless Area of western Wisconsin, two effigy mound societies
have been recognized on the basis of discrete mound types, ceramics, and projectile
points. Late Woodland sites in the northern portion of this region contain long-tailed
(“panther/turtle”) quadruped mounds, Angelo Punctated pottery, and unnotched
Madison Triangular points made from locally available silicified sandstone, which
are collectively identified with the Lewis phase. To the south, in what is called the
Eastman phase, most quadruped mounds have no tails (“bears”) or relatively
short tails (“deer, canines,” etc.), the ceramics are dominated by various Madisonware types, and projectile points include both unnotched and side-notched forms,
nearly all of which are made of local chert (Boszhardt and Goetz 2000; Theler
and Boszhardt 2003, 2006). The earliest evidence of Middle Mississippian contact
in the UMRV are at the Fisher Mounds Site Complex (47Ve825), at the mouth of
Coon Valley and in an apparent buffer zone between Lewis and Eastman phase territories (Arzigian 2008; Benden 2004; Boszhardt et al. 2010), and a complex consisting of Squier Garden (47Tr156), Uhl (47Tr159), Pelkey (47Tr415), and the Little
Bluff platform mounds (47Tr32) at the Trempealeau Bluffs (Benden et al. 2011;
Boszhardt and Benden 2014; Boszhardt et al. 2012, 2013, 2015; Green and
Rodell 1994). These early Mississippian components represent short-lived colonies
PETROGRAPHIC ANALYSIS OF LATE WOODLAND AND MIDDLE MISSISSIPPIAN CERAMICS
3
whose assemblages are dominated by lithics and ceramic vessels that were imported
from the south. On the basis of high frequencies of red- and dark-slipped vessels with
extruded rather than rolled lips, the complete absence of Ramey Incised, a tight suite
of radiocarbon dates, and architectural characteristics, the Fisher and Trempealeau
colonies correspond with the ca. A.D. 1050 Edelhardt/Lohmann phase transition in
the American Bottom (Pauketat et al. 2015; Stoltman et al. 2008). Fisher and Trempealeau represent the only confirmed examples of pre-Stirling Mississippian
expressions to the north of Cahokia, although Aztalan in southeastern Wisconsin
may have a minor Lohmann phase component (Richards 2007).
Later Middle Mississippian influence in the upper Midwest in the eleventh and
twelfth centuries is indicated by a variety of archaeological manifestations between
Lake Michigan and the central Missouri River (Hall 1991; Kelly 1991a, 1991b;
Milner 1991; Stoltman 1991a, 2000). Examples of Middle Mississippian interaction
in the UMRV include Bennett phase sites along the lower Apple River (Emerson 1991;
Emerson et al. 2007; Millhouse 2012), the Fred Edwards site in southwest Wisconsin
(Finney 2013; Finney and Stoltman 1991), Hartley Fort in northeast Iowa (Finney
1993; Finney et al. 1993; McKusick 1964; Tiffany 1982), and Silvernale phase
sites in the Red Wing locality (Gibbon 1979; Gibbon and Dobbs 1991; Rodell
1991). These and numerous other sites indicate a surge of Mississippian influences
to the northern hinterland after A.D. 1100 (Figure 1). They include imported or
figure 1 Map of Driftless Area showing major northern frontier sites.
4
ROBERT F. BOSZHARDT AND JAMES B. STOLTMAN
local imitations of the Stirling phase diagnostic pottery type Ramey Incised and thus
are affiliated with the Ramey horizon (ca. A.D. 1100–1200) (Green 1997; Hall 1991;
Overstreet 2000). Fred Edwards and Hartley Fort were palisaded villages that contained a blend of Terminal Late Woodland and Middle Mississippian pottery (Stoltman 1991b) and likely date to the first half of the Stirling phase (ca. A.D. 1100–1150).
The main Mississippian components at Apple River and Red Wing extend later in
time (A.D. 1150–1200) and blend into the early manifestations of the Oneota
culture (Emerson et al. 2007; Millhouse 2012).
In 2002 and 2003, salvage excavations were conducted at the Iva site (47Lc42) on
the outskirts of Onalaska, Wisconsin (Boszhardt 2004). That work identified a mixed
Terminal Late Woodland and Middle Mississippian component that also corresponds
to the Ramey horizon. Several features produced diagnostic Angelo Punctated sherds
and portions of Ramey Incised and late variety (rolled rim) Powell Plain vessels.
Angelo Punctated is the type attributed to Terminal Late Woodland occupants of
the northern portion of the Driftless Area (Boszhardt 1996; Boszhardt and Goetz
2000). The presence of Angelo Punctated in direct association with Ramey Incised
and Powell Plain demonstrates Ramey horizon interaction at Iva. One basin
(Feature 16) produced sections of Mississippian fine-ware vessels along with Angelo
Punctated sherds and associated faunal and floral remains that suggest a feasting
event with representatives of both societies participating (Boszhardt 2004:74–76).
How Middle Mississippian and Late Woodland people made contact with and
influenced one another in the UMRV are ongoing research questions. Stoltman
(1991a, 2000) provided a series of five Culture Contact Situation (CCS) models
as a framework for evaluating degrees of interaction for particular Middle Mississippian components in the UMRV. In addition, Pauketat and Emerson (1991) have
suggested that Ramey Incised vessels might have provided a vehicle for sharing
Middle Mississippian ideologies through ritual ceremonies. The discovery of a
Middle Mississippian component at the Iva site in the La Crosse locality provides
an opportunity to assess both Stoltman’s contact situation models and the “power
of the pot” exchange concept of Pauketat and Emerson.
Boszhardt (2004) described the details of the 2002–2003 salvage excavations at
Iva and offered interpretations for cultural sequencing, interaction models, feasting,
and potential disease factors. Subsequent to that report, Stoltman conducted petrographic analyses of 13 Iva vessels, five of which are typologically Late Woodland
and eight of which conform to Mississippian wares. The results provide objective
data on the representation of Mississippian interaction in the ceramic assemblage.
This paper provides a condensed review of the Iva site with a focus on the Late
Woodland and Mississippian ceramic assemblages and presents the results of the
petrographic analyses along with a discussion of the implications at a regional level.
Environmental setting
The Iva site is situated within the northern portion of Wisconsin’s Driftless Area (see
Figure 1). This unglaciated region is rugged, with narrow, deeply incised drainages
that are separated by dendritic upland ridges (Martin 1965; Mickelson et al. 1981).
Periodic glacial meltwater floods affected Driftless Area drainages and left a series of
PETROGRAPHIC ANALYSIS OF LATE WOODLAND AND MIDDLE MISSISSIPPIAN CERAMICS
5
sand and gravel terraces along the Upper Mississippi River (Flock 1983; Knox 1996;
Martin 1965:156–168).
The underlying regional bedrock is primarily Cambrian sandstone capped by formations of Ordovician dolomitic limestone. The pre-1850 vegetation of the Driftless
Area was a mosaic of prairie, savanna, and forest (Finley 1976). In general, prairies
and oak savannas occupied sandy terraces, southwest-facing bluffs, and open
upland ridgetops, all of which were susceptible to fire. Forested areas included
gallery strips along floodplains, including islands in the Mississippi River bottom,
and fire-protected northeast-facing slopes (Theler and Boszhardt 2003).
Specifically, the Iva site is located at the mouth of Sand Lake Coulee, a small (2.5
km long) valley along the Wisconsin bluff line and 5 km east of the Mississippi River
floodplain (Figure 2). The small, perennial stream that drains Sand Lake Coulee was
blocked from reaching the Mississippi River by the Onalaska Terrace and instead
fertilized a several-hundred acre swale between the base of loess-capped bluffs
and the dune-modified Onalaska Terrace (Boszhardt 1985a, 1985b, 2009). This
swale supported a wet prairie that formed a deep rich black soil, which is ideal
for growing corn. The Iva site occupied a west-facing slope that descended from
figure 2 Map of Sand Lake Coulee showing the location of the Iva site and the Sand Lake
Archaeological District.
6
ROBERT F. BOSZHARDT AND JAMES B. STOLTMAN
the rolling sand dunes on the adjacent Krause site (47Lc41) to the bottomland soil of
the Sand Lake Creek floodplain.
Although the Onalaska Terrace is dominated by sand and gravel outwash, clays
and fine silts for making ceramics are available from four local settings. Red ironrich clays exist in the decomposing limestone bluff tops and often contain frostfractured Prairie du Chien chert inclusions (Frolking 1982). At the base of the
bluffs, alluvially redeposited (yellow-brown) loess accumulations exist that
nineteenth-century settlers found suitable for manufacturing bricks (Buckley
1901:160–176). Silts and clays (gray to black) also exist in local streambeds, such
as Sand Lake Coulee, and in the alluvial floodplains of the Mississippi to the
west. The finest clays in the area are red-pink colored deposits representing slackwater sediments from Glacial Lake Superior outwash flood events (Flock 1983).
Near-surface deposits of these pure clays would have been available to prehistoric
potters at the mouths of several local valleys (Ketterhagen 2006), including Sand
Lake Coulee and Long Coulee, and in the lower La Crosse River valley (47Lc71).
Previous investigations
The Iva site was first reported during a 1979 surface collection of a 13-acre field
during a survey of Sand Lake Coulee that produced no Mississippian diagnostics
(Gallagher 1980:8–9). No further investigations were undertaken at Iva until the
fall of 2002, when the site was slated for residential development. However, extensive archaeological work at adjacent sites at the mouth of Sand Lake Coulee between
1982 and 2009 recovered evidence of a major Oneota settlement and associated
ridged field complexes (Boszhardt et al. 1985; Gallagher et al. 1985, 1987; Sasso
et al. 1985). In addition, minor Middle and Late Woodland components were documented (Boszhardt 1985b). Early recognition of the significance of the Oneota
complex here resulted in listing the Sand Lake Archaeological District on the
National Register of Historic Places in 1984 (see Figure 2), with Iva included as a
contributing resource (Stevenson and Boszhardt 1983).
In 1986, excavations were undertaken on a sand knoll at the Krause site (47Lc41),
about 100 m north of the Iva field, under the direction of James Theler. The results of
those investigations have not yet been formally written up, but several recovered rim
sherds indicate a minor Middle Mississippian component (Boszhardt 1996:135,
1997:145). At that time, evidence of Middle Mississippian activity in the La
Crosse locality was rare (Boszhardt 1989:64; Rodell 1989).
In 1988, the Cedar Creek Golf Course was constructed over the Sand Lake site
(47Lc44), and since the late 1990s, several subdivisions have been developed
within the Sand Lake Archaeological District. For example, between 2000 and
2009, the Krause, Meier Farm (47Lc432), Lower Sand Lake (47Lc45), and
Northern Engraving sites (47Lc164) were all transformed from farm to residential
use. Extensive data-recovery excavations were conducted at all of these sites
(Boszhardt 2009; Boszhardt and Holtz-Lieth 2008). From these investigations,
the only other Middle Mississippian sherds found in the Sand Lake Archaeological District are 11 from the Lower Sand Lake site (47Lc45), located immediately
PETROGRAPHIC ANALYSIS OF LATE WOODLAND AND MIDDLE MISSISSIPPIAN CERAMICS
7
south of Iva (Boszhardt and Holtz-Lieth 2008:7–17). Based on the known distribution of Middle Mississippian ceramics in Sand Lake Coulee, the associated
component appears to encompass approximately 200 × 100 m and is centered
on the Iva site.
2002–2003 excavations at Iva
In the fall of 2002, Boszhardt was notified of an impending condominium development with a cul-de-sac and retention pond that would affect the Iva site. The private
development did not require regulatory compliance, and all related archaeological
efforts were voluntary. A bulldozer was used to scrape the historic plow zone
from 37 trenches, which exposed 30 features. Some of these, particularly at the
north end, were typical post-A.D. 1300 Oneota storage refuse pits; however,
several hearth-like features in the central area represented earlier Late Woodland
and Middle Mississippian activities (Figure 3). The dozer backdirt piles were collected multiple times, resulting in the recovery of diagnostic Middle Woodland,
Late Woodland, Middle Mississippian, and Oneota pottery sherds, as well as a
broken ear spool made of Baraboo pipestone. The following provenience summaries
figure 3 Map showing excavation units with locations of Features 15, 16, 17, and X and
Dozer Scrapes 16 and 17 at the Iva site.
8
ROBERT F. BOSZHARDT AND JAMES B. STOLTMAN
TABLE 1
LITHIC TOOLS AND DIAGNOSTIC MIDDLE MISSISSIPPIAN AND LATE WOODLAND POTTERY
Surface
Points
End scrapers
Modified flakes
Middle Mississippian Pottery
Late Woodland Pottery
6
F.15
F.16
F.X
5N
10W
0
0
0
0
5N
12W
5N
14W
5N
16W
7N
8W
7N
10W
7N
12W
Total
1
0
0
1
0
0
8
6
1
1
1
0
1
0
0
0
0
0
10
23
0
1
2
0
5
0
0
0
0
0
31
28
3
12
13
2
13
9
6
3
1
0
90
126a
1
12
4
3
3
17
2
7
37
1
213
a
Only three are Madison ware and one is Aztalan Collared. The majority of the other decorated Late Woodland sherds
are Angelo Punctated.
review associated ceramic assemblages and include references to Stoltman’s thin
section codes (e.g., No. 47-XXX).
Surface collections: Most of the Late Woodland diagnostics from the backdirt
piles (see Table 1) are Angelo Punctated sherds (Figure 4). Madison ware is represented by a single simple cord-twist Madison Cord-Impressed neck sherd
(Figure 5, No. 47-336) and three small “Madison Plain” rims. In addition, a
single Aztalan Collared rim (No. 47-330) was recovered (Figure 6). Aztalan Collared is well dated between ca. A.D. 1050 and A.D. 1150 and is most common in
glaciated southeastern Wisconsin, sporadic in the southern Driftless Area, and
exceedingly rare in the northern Driftless Area (Kelly 2003). The Aztalan Collared
rim from Iva may represent the northernmost known example of this type along the
UMRV. Middle Mississippian ceramics from the surface include both shell- and grittempered rolled rims; several burnished, shell-tempered body sherds; and a few redslipped, shell-tempered sherds.
Middle Mississippian–Late Woodland features
Several features in the central portion of the site contained Late Woodland and/or
Middle Mississippian pottery. Table 1 summarizes the lithic tools and diagnostic
Late Woodland–Middle Mississippian sherd tallies from these features and encompassing excavation units. For a more complete discussion of the feature sizes, forms,
lithics, and associated flora and fauna, see Boszhardt (2004).
Feature 15 included 13 cord-roughened body sherds, all of which are relatively
small and thin with fine grit-temper particles, which suggest Late Woodland affinity.
One grit-tempered sherd is decorated with tool impressions over a smooth surface,
implying an intrusive Middle Woodland fragment. The largest sherd in the feature
assemblage is a shell-tempered body sherd that is cord roughened.
Feature 16 produced two Angelo Punctated vessels that are represented by rims
(see Figures 4a and 7). One (Figure 7) is a variant decorated with trailed lines and
oval punctates rather than the more common fine incised lines and angular punctations. In addition, this vessel has tool impressions along the inner lip margin,
PETROGRAPHIC ANALYSIS OF LATE WOODLAND AND MIDDLE MISSISSIPPIAN CERAMICS
9
figure 4 Angelo Punctated sherds from Iva: (a) rim from F16; (b) rim from Unit 5N, 10W; (c)
rim from Unit 5N, 12W; (d) rim from Unit 5N, 14W; (e) rim from Unit 5N, 16W; (f) rim from Unit
7N, 10W; (g) rim from Trench 17; (h) one of two angular shoulders from surface.
whereas most Angelo Punctated vessels have sharp notching on the lip top (Boszhardt 1996).
Three Mississippian vessels are represented by shell-tempered rims from Feature
16. Two are Ramey Incised jars. The first (No. 47-327) is a small rolled-lip jar
with a sharp angular shoulder and a double-zigzag line inscribed on the shoulder
(Figure 8a). The second (No. 47-325) has a burnished rolled rim and an angular
shoulder on which was engraved a distinct “forked-eye” motif (Figure 8b). The
second jar also exhibits a ring of red pigment along the interior lip. The third Mississippian vessel (No. 47-336) from Feature 16 is a bowl that has a tan slip and
10
ROBERT F. BOSZHARDT AND JAMES B. STOLTMAN
figure 5 Two Madison Cord-Impressed decorated body sherds, thin section numbers
47-336 and 47-337.
figure 6 Aztalan Collared (T.S. No. 47-330) and two grit-tempered, cordmarked vessels
with Powell-like angular shoulders (T.S. Nos. 47-331 and 47-332).
burnished exterior surface (Figure 9). The context of Feature 16 during excavation
makes it clear that the large Angelo Punctated rim section was directly associated
with the three Mississippian rims.
Feature 17 consisted of a layer of burned sandstone slabs with a few burnished
shell-tempered sherds nearby. The burnished sherds articulate with others from
Feature 16 to form an angular shoulder of a bowl. Feature X contained several
Middle Mississippian rims that were mixed with Woodland and Oneota sherds
from extensive rodent churning. The Late Woodland sherds from Feature X
include a few Angelo Punctated fragments, one of which is from the large vessel
in Feature 16. The Middle Mississippian ceramic assemblage from this feature consists of four rolled rims, one of which exhibits parallel curvilinear trails that correlate
to a variant of the Ramey Incised type, several black burnished sherds (including one
angular shoulder fragment), and two red-slipped sherds.
PETROGRAPHIC ANALYSIS OF LATE WOODLAND AND MIDDLE MISSISSIPPIAN CERAMICS
11
figure 7 Angelo Punctated rim (T.S. No. 47-291) from Feature 16.
figure 8 Two Ramey Incised vessels from Feature 16: (a) suspected of being locally made
(T.S. No. 47-327); (b) suspected of being an import (T.S. No. 47-325).
12
ROBERT F. BOSZHARDT AND JAMES B. STOLTMAN
figure 9 Five shell-tempered vessels identified by thin section number (see Table 4).
Late Woodland–Middle Mississippian diagnostics from excavated
units
Diagnostic Late Woodland and Middle Mississippian sherds were also recovered
from excavation of grid units that were established in order to clarify the boundaries of select features and obtain a sample of the occupation area between features
(see Figure 3). The units revealed general midden dispersal of some vessel fragments.
For example, sherds from the large Angelo Punctated section in Feature 16 were
found immediately outside the basin in adjacent Units 5N, 10W; 7N, 8W; and
7N, 10W, as well as up to 4 m away in Units 5N, 14W and 7N, 12W, the latter
of which was adjacent to Feature X, which also produced a sherd from this
vessel. There is no indication that the relatively large section of the Angelo Punctated vessel and the three Mississippian rims were mixed into Feature 16 by later
disturbances.
Unit 5N, 10W encompassed the south half of Feature 16. Excavation outside the
feature produced an Angelo Punctated rim and decorated body sherd and an angular
shoulder of a shell-tempered vessel. The Angelo body sherd is from the same vessel
as the large rim and shoulder within Feature 16, but it does not articulate. The rim
(see Figure 4b) is part of a different Angelo Punctated vessel. The angular shoulder
sherd represents a Mississippian vessel form.
Unit 5N, 12W produced a small Angelo Punctated rim (see Figure 4c) and an
angular shoulder of a shell-tempered Middle Mississippian vessel.
Unit 5N, 14W produced several Late Woodland and Middle Mississippian diagnostics. These include two rims from different Angelo Punctated vessels (see
Figure 4d). One of these may be from the same vessel as the rim recovered in Unit
5N, 10W. In addition, two Angelo Punctated body sherds were recovered, one of
which is from the large Feature 16 vessel. One of the few Madison Cord-Impressed
sherds found at the site came from this unit. This sherd (No. 47-337) is from the neck
area (the lip is missing) and is decorated with horizontal cord impressions over a
cord-roughened surface (see Figure 5). In addition, two burnished shell-tempered
sherds were found during the excavation of Unit 5N, 14W.
PETROGRAPHIC ANALYSIS OF LATE WOODLAND AND MIDDLE MISSISSIPPIAN CERAMICS
13
Unit 5N, 16W encompassed Feature 13, an Oneota pit with a few earlier Late
Woodland sherds that were probably mixed in fortuitously. Excavation adjacent
to the feature recovered a single Angelo Punctated rim sherd (see Figure 4e).
Unit 7N, 8W was situated immediately east of Feature 16. Excavation of the unit
recovered a small rolled rim of a black burnished vessel and a red-slipped body
sherd. Both are shell tempered and represent Middle Mississippian affiliation. The
unit also contained a decorated Angelo Punctated body sherd that is from the
large vessel associated with Feature 16.
Unit 7N, 10W encompassed the north half of Feature 16. The unit adjacent to the
feature produced several Late Woodland and Middle Mississippian diagnostics.
These include a rim (see Figure 4f) and two body sherds from Angelo Punctated
vessels. One of the body sherds is from the large vessel associated with Feature
16. In addition, two burnished, shell-tempered sherds were recovered, one of
which has a red slip. Another recovered shell-tempered sherd has a cord-roughened
surface, similar to the one found in Feature 15. These are likely affiliated with a
mixed Late Woodland–Mississippian component ca. A.D. 1100–1150.
Unit 7N, 12W overlapped the southeastern portion of Feature X. The unit outside
the feature produced a body sherd from the large Angelo Punctated vessel associated
with Feature 16.
Finally, a grit-tempered vessel portion with a distinct angular shoulder was found
on the scraped floor of Dozer Trench 14. Unfortunately, the bulldozer hit this vessel,
and the rim is missing. Nonetheless, the sharp angular shoulder is distinctly Middle
Mississippian, revealing another example of blended Woodland and Mississippian
ceramic attributes in one vessel.
Petrography
Thin sections were prepared for 13 pottery vessels recovered at the Iva site and analyzed by Stoltman for quantifiable petrographic attributes (Table 2). The analytical
procedure utilized, referred to as point counting, is described in Stoltman (1989,
1991b, 2001). Petrography when applied to pottery has some important advantages
and limitations that merit discussion. When viewing a pottery thin section, one sees a
variety of discrete mineral grains of silt size (0.002–0.0624 mm) and larger that are
encompassed within an amorphous matrix, usually brownish or reddish in color.
The matrix consists of various clay minerals, none of which can be identified specifically because of their extremely fine sizes. The coarser mineral grains, especially
those the size of sand (0.0625–1.99 mm) and gravel (>2.00 mm), by contrast, are
the primary focus because they appear as discrete particles whose mineralogical
compositions can be determined with a high degree of reliability.
Typically, pottery matrix is composed of “clays” to which aplastic “tempers” have
been added by the pot makers. Petrography offers a unique approach to ceramic
analysis in that it allows reliable identification of not only the humanly added
tempers but also the clay sources used by the potters. This hinges on the fact that
few natural clay-rich sediments are pure. Rather, most clays, depending on their geological history, are composed of varying amounts of natural inclusions of silt and
14
ROBERT F. BOSZHARDT AND JAMES B. STOLTMAN
TABLE 2
THIN-SECTIONED SHERDS FROM IVA
Thin section number
Provenience
Temper
Pottery type
Figure #
47-330
Surface
Grit
Aztalan Collared
7
47-271
Surface
Grit
Angelo Punctated
–
47-291
Feature 16
Grit
Angelo Punctated
8
47-336
Surface, Trench 17
Grit
Madison Cord-Impressed
6
47-337
5N 14W, Level 6
40–50 cm BD
Grit
Madison Cord-Impressed
6
47-331
Surface
Grit
Powell-like
7
47-332
Surface
Grit
Powell-like
7
47-325
Feature 16
Shell
Ramey Incised
9b
47-328
Dozer Trench 17
Shell
Ramey Incised
10
47-329
Surface
Shell
Ramey Incised
10
47-327
Feature 16
Shell
Ramey Incised
9a
47-326
Feature 16
Shell
Bowl
10
47-333
Surface
Shell
Bottle
10
sand (as used here, these are purely size, not mineralogical, designations) along with
the clay minerals. Petrography, by discriminating the temper, enables analysts to
characterize clayey matrices in pottery thin sections in terms of the relative proportions of silt, sand, and clay and, by so doing, offers the potential to identify distinctive sources of clay-rich sediments used by the potters.
In the current study, two temper types were identified: grit (granitic rock in this
case) (Figures 10 and 11) and shell (Figures 12 and 13). In the ensuing discussion,
this temper distinction provides the basis for distinguishing two pottery groups. In
considering these two groups, two parameters of composition — body and paste
— will be employed. These are recorded in Tables 3 and 4 and shown graphically
in Figures 14–17.
The term body means the bulk composition of a vessel and is characterized by the
relative proportions of temper, clay matrix, and natural sand-size grains in the
matrix (see Table 3). Silt particles — that is, visible mineral grains smaller than
0.0625 mm in maximum diameter — are difficult to identify mineralogically and,
for the most part, can be accepted as natural inclusions. Thus, silt grains are
recorded as part of the “matrix” in the body index. An exception in this regard pertains to shell: silt-size shell grains could be reliably identified and, as demonstrable
human additives, were recorded as temper. The body index reflects customary and
presumably intentional human actions in selecting and mixing ingredients to
fashion serviceable pottery containers. Insofar as this index can be accepted as an
embodiment of such past human behavior, it has the potential to be informative
in several ways. For example, temper types, amounts, and sizes may reflect regional,
temporal, and/or functional patterns in the practices of past potters.
PETROGRAPHIC ANALYSIS OF LATE WOODLAND AND MIDDLE MISSISSIPPIAN CERAMICS
15
figure 10 Photomicrograph of thin section No. 47-331 from cordmarked, granite-tempered
Powell-like vessel taken at 10× magnification under crossed polars.
Paste, by contrast, refers to the composition of the clay-rich sediments from which
pottery vessels were made, excluding the temper. In Tables 3 and 4, paste is
expressed as the relative percentages of matrix (i.e., clay), silt, and sand. These
values have the potential of being able to discriminate residual from alluvial from
figure 11 Photomicrograph of thin section No. 47-336 from granite-tempered Madison
Cord-Impressed vessel taken at 10× magnification under crossed polars.
16
ROBERT F. BOSZHARDT AND JAMES B. STOLTMAN
figure 12 Photomicrograph of thin section No. 47-325 from imported, shell-tempered
Ramey Incised vessel taken at 10× magnification under crossed polars. Note relative abundance of shell and paucity of sand and silt.
aeolian clay-rich sediments, thus enabling possible identification of the potter’s clay
sources.
One other parameter, size of mineral inclusions larger than silt, was recorded for
each thin section. Rather than measure each sand-size grain directly, each such grain
figure 13 Photomicrograph of thin section No. 47-327 from locally made, shell-tempered
Ramey Incised vessel taken at 10× magnification under crossed polars. Note relative
paucity of shell and abundance of sand and silt compared to Figure 12. The 0.80 mm identifies a sandstone fragment.
PETROGRAPHIC ANALYSIS OF LATE WOODLAND AND MIDDLE MISSISSIPPIAN CERAMICS
17
TABLE 3
BODY AND PASTE VALUES FOR GRIT-TEMPERED VESSELS FROM THE IVA SITE
Type
Thin section no.
%Matrix
%Sand
%Temper
Temper type
Temper size index
Aztalan Collared
47-330
86
4
10
Granite
3.77
Angelo Punctated
47-271
72
1
27
Granite
3.22
Angelo Punctated
47-291
71
3
26
Granite
3.36
Madison Cord-Impressed
47-336
75
6
19
Granite
3.52
Madison Cord-Impressed
47-337
79
2
19
Granite
3.46
Cordmarked, Powell-like
47-331
82
0
18
Granite
2.86
Cordmarked, Powell-like
47-332
82
0
18
Granite
2.87
Thin section no.
%Matrix
%Silt
%Sand
Sand-size index
Aztalan Collared
47-330
92
4
4
Angelo Punctated
47-271
97
2
1
1.00
Angelo Punctated
47-291
91
5
4
2.00
Madison Cord-Impressed
47-336
81
11
8
1.54
Madison Cord-Impressed
47-337
94
4
2
1.00
Cordmarked, Powell-like
47-331
97
3
0
–
Cordmarked, Powell-like
47-332
97
3
0
–
Body
Type
Paste
1.57
encountered during the analysis was assigned a value from 1 to 5 based on the following ordinal scale:
1 = Fine
0.0625–0.249 mm
2 = Medium
0.25–0.499 mm
3 = Coarse
0.50–0.99 mm
4 = Very coarse
1.00–1.99 mm
5 = Gravel
2.00+ mm
The individual grain-size values were then summed and divided by the total
number of sand and gravel grains counted, thus providing a mean “sand-size
index,” ranging between 1 and 5, for each thin section. For body, this value pertains
to temper, and for paste, it pertains to natural inclusions the size of sand and larger.
Grit-tempered vessels
Seven of the thin-sectioned vessels were tempered with rocks of granitic composition
(e.g., LeMaitre 2002) (see Figures 10 and 11). A bedrock source for the granite does
not exist locally, making local glacial outwash its most likely source. On stylistic
grounds, five of these vessels conform to three local Late Woodland types:
Madison Cord-Impressed (n = 2; see Figure 5), Angelo Punctated (n = 2; see
18
ROBERT F. BOSZHARDT AND JAMES B. STOLTMAN
TABLE 4
BODY AND PASTE VALUES FOR SHELL-TEMPERED VESSELS FROM THE IVA SITE
Type
Thin section no.
%Matrix
%Sand
%Temper
Temper type
Temper size index
Body
Ramey Incisedb
47-325
79
0
21
b
Ramey Incised
47-328
69
0
Ramey Incised
47-329
82
3
Shell
2.00
31
Shell
2.00
15
aShell
2.74
Ramey Incised
47-327
77
10
13
aShell
2.77
Bowlb
47-326
68
3
29
Shell
2.48
Bottleb
47-333
2.10
Type
Thin section no.
79
1
20
aShell
%Matrix
%Silt
%Sand
Sand-size index
Paste
Ramey Incisedb
47-325
99
1
0
–
Ramey Incisedb
47-328
97
3
0
–
Ramey Incised
47-329
92
4
4
1.60
Ramey Incised
47-327
79
10
11
1.24
b
Bowl
47-326
93
3
4
1.00
Bottleb
47-333
94
5
1
1.00
a
Leached shell.
Vessels with slipped surfaces.
b
Figure 7), and Aztalan Collared (n = 1; see Figure 6). The remaining two, while cordmarked and grit tempered, have sharp-shouldered vessel forms reminiscent of the
Mississippian Powell Plain and Ramey Incised (see Figure 6) types. Such vessels
are known from other sites in Wisconsin bearing unmistakable evidence of
Middle Mississippian influences (e.g., Aztalan and Fred Edwards), where these
pots are viewed as hybrid Woodland–Mississippian types (Finney and Stoltman
1991; Richards 2007). The compositions of these vessels are recorded in Table 3
and presented graphically in Figures 14 and 15, with a thin-section view of
47-331 in Figure 10.
Both the stylistic and compositional properties of these seven vessels are consistent
with the view that all were manufactured locally — that is, within the upper Mississippi Valley region. Determining whether all were manufactured at the Iva site is
more problematic. Figure 14 and Table 3 show that the paste values for six of the
seven vessels are similar enough (all possessing sand and silt values of 5 percent
or less) to be considered derived from the same, or closely similar, clays. One potentially significant distinction within these data pertains to the two hybrid vessels, both
of which were made from virtually sand-free clays (e.g., see Figure 14). Within such a
small sample, it seems prudent to view these data as suggestive but not conclusive
evidence of a separate origin for the two hybrid vessels. Meanwhile, it is safe to conclude that the most likely source for the low-silt, low-sand pastes that characterize
six of the seven grit-tempered vessels would have been slack-water sediments
PETROGRAPHIC ANALYSIS OF LATE WOODLAND AND MIDDLE MISSISSIPPIAN CERAMICS
19
figure 14 Ternary graph showing the paste values for seven grit-tempered vessels from Iva.
figure 15
Ternary graph showing the body values for seven grit-tempered vessels from Iva.
20
ROBERT F. BOSZHARDT AND JAMES B. STOLTMAN
figure 16
Ternary graph showing the paste values for six shell-tempered vessels from Iva.
within the Mississippi alluvial valley (see local clay source summary in “Environmental setting” section).
The seventh grit-tempered vessel (No. 47-336), one of the two Madison
Cord-Impressed vessels (see Figure 5), is compositionally distinct in being
figure 17
Ternary graph showing the body values for six shell-tempered vessels from Iva.
PETROGRAPHIC ANALYSIS OF LATE WOODLAND AND MIDDLE MISSISSIPPIAN CERAMICS
21
characterized by uniquely high silt (11 percent) and sand (8 percent) values (see
Table 3 and Figure 11). Whether this reflects a different source or merely an extension of the range of variation of local raw materials is currently unknown. A larger
sample is needed to resolve this issue.
When considering the body values (see Table 3 and Figure 15), much more
internal diversity is evident in contrast to the paste values (cf. Figure 14 and
Table 3). Moreover, an interesting tendency for vessels of the same type to cluster
seems to be present (see Figure 15). The two Angelo Punctated vessels have the greatest temper volume at 26 and 27 percent, while the Madison Cord-Impressed and
hybrid vessels all have 18–19 percent temper (see Figure 15). The Aztalan Collared
vessel stands out in having the lowest temper volume at 10 percent and also the coarsest temper of the seven grit-tempered vessels with a size index of 3.77 (see Table 3).
Once again, the close similarity of the two hybrid vessels is striking, with both
having 18 percent temper. In addition, the two have the finest temper among the
seven vessels, with size indices under 2.90 (see Table 3).
From these compositional data, a number of inferences are possible. All seven
vessels were apparently made from alluvial clays tempered with granitic rock.
This implies a generally common origin, at least within the UMRV. The body
indices — both the amount and size of temper — however, vary in seemingly nonrandom ways. These data are suggestive of the presence of multiple potters at the
site, each making use of distinctive recipes in fabricating his or her own particular
pottery style, although a nonlocal origin of some of these vessels (especially the
Aztalan Collared and Madison Cord-Impressed) is a credible possibility.
Shell-tempered vessels
Six of the thin-sectioned vessels are shell tempered. These consist of four Ramey
Incised jars, one bowl, and one bottle; all of these types are closely associated
with Mississippian culture in the American Bottom (e.g., Bareis and Porter 1984).
The compositional values for these vessels are recorded in Table 4 and portrayed
graphically in Figures 16 and 17.
From both the paste and body indices, it can be seen that the Ramey Incised jars
form two distinct compositional subgroups (see Table 4; see Figures 16 and 17). One
subgroup (composed of vessels 47-325 and 47-328) has a paste with virtually no
sand and only 1–3 percent silt (see Figure 16); these same two vessels have bodies
with abundant shell temper (21–31 percent) that is relatively fine (size indices of
2.00) (see Table 4; see Figure 17). In contrast, the second subgroup (composed of
vessels 47-327 and 47-329) has siltier (4–10 percent) and sandier (4–11 percent)
pastes (see Figure 16); the amount of shell temper in this subgroup is notably less
(13–15 percent) and coarser (size indices of 2.74 and 2.77) compared to the other
subgroup (see Table 4 and Figure 17).
By comparing the compositions of the former two Ramey Incised vessels to thinsection data from a sample of Ramey Incised vessels from the American Bottom, it
can be seen that an American Bottom origin for the first subgroup of Iva vessels
(47-325 and 47-328) can be reasonably postulated. The sample of 25 Ramey
22
ROBERT F. BOSZHARDT AND JAMES B. STOLTMAN
Incised vessels from the American Bottom derives from the following sites: Tract 15
A (n = 5) and the Dunham Tract (n = 3) at Cahokia (Pauketat 1998); Sponemann (n
= 8) (Jackson et al. 1992); BBB Motor (n = 4) (Emerson and Jackson 1984);
Rapps-Lizard (n = 1) (Pauketat 1984); and Range (n = 4) (Kelly 1990). The means
and standard deviations for paste and body for these 25 vessels are presented in
Table 5. As summarized in this table, the salient properties of Ramey Incised
vessels from the American Bottom are (1) a low-silt, low-sand paste; (2) shell
temper that typically exceeds 20 percent of the body; and (3) relatively finely
crushed shell — that is, a size index typically <2.50. In addition, American
Bottom vessels invariably have slipped surfaces.
As can be seen from Tables 4 and 5, two of the Iva Ramey Incised vessels (Nos.
47-325 and 47-328) have both paste and body values that fall fully within the
range of variation of the American Bottom sample, and both vessels have slipped
surfaces. It is reasonable to view these vessels as American Bottom imports. By contrast, the other two Ramey Incised vessels (Nos. 47-327 and 47-329) are distinctly
different in all three of the properties that characterize American Bottom vessels.
That is, these two vessels have pastes with higher silt and sand values, bodies with
notably lower temper percentages, and coarser shell temper (see Tables 4 and 5).
Moreover, neither of these vessels is slipped. Considering these factors, it is reasonable to regard vessels 47-327 and 47-329 as locally produced rather than imported
from the American Bottom (cf. Figures 15 and 16).
The bowl (47-326) and the bottle (47-333) are noteworthy as typical Middle Mississippian vessel forms that are rare in the upper Mississippi Valley region, with the
notable exception of Aztalan (Barrett 1933; Richards 1992). Indeed, so unusual are
these vessel forms in the UMRV that on that basis alone their importation from the
TABLE 5
MEANS AND STANDARD DEVIATIONS OF PASTE AND BODY VALUES FOR 25 RAMEY INCISED VESSELS FROM
SIX AMERICAN BOTTOM SITES VS. THE SIX SHELL-TEMPERED IVA VESSELS
Sites/types
N
%Matrix
%Silt
%Sand
Sand-size index
Paste
American Bottom, Ramey Incised
25
96.0 ± 2.8
3.3 ± 2.7
0.7 ± 0.7
1.17 ± .36
Iva, imported Ramey Incised
2
98.0 ± 1.4
2.0 ± 1.0
0±0
–
Iva, local Ramey Incised
2
85.5 ± 9.2
7.0 ± 4.2
7.5 ± 5.0
1.42 ± 0.26
Bowl
1
93
3
4
1.00
Bottle
1
94
5
1
1.00
Sites/types
N
%Matrix
%Sand
%Temper
Temper size index
25
78.1 ± 6.5
0.6 ± 0.6
21.3 ± 6.4
2.24 ± 0.37
Body
American Bottom, Ramey Incised
Iva, imported Ramey Incised
2
74.0 ± 7.1
0
26.0 ± 7.1
2.00 ± 0
Iva, local Ramey Incised
2
79.5 ± 3.5
6.5 ± 5.0
14.0 ± 1.4
2.76 ± 0.02
Bowl
1
68
3
29
Shell
Bottle
1
79
1
20
Shell
PETROGRAPHIC ANALYSIS OF LATE WOODLAND AND MIDDLE MISSISSIPPIAN CERAMICS
23
south can be postulated. It is also noteworthy that both vessels have slipped surfaces.
In comparing the compositions of these vessels with the Ramey Incised sample from
the American Bottom, a caveat should be mentioned; namely, bowls and bottles may
differ in composition from Ramey Incised jars because of functional considerations
on the part of the makers or due to the limited/specialized production of the Ramey
jars (e.g., Pauketat and Emerson 1991). With this caveat in mind, it can be seen in
Table 5 that all values of the body and paste indices for these vessels closely resemble
those of the two presumed imported Ramey Incised vessels, as well as the Ramey
Incised sample from the American Bottom. It might be noted that the bowl
(47-326) has a paste that is virtually identical to that of Ramey Incised vessel No.
47-329, which has been identified as of probable local origin. While a local origin
of the bowl thus cannot be completely discounted, considering that all its paste
and body values, except the 4 percent sand value for paste, fall within two standard
deviations of the American Bottom mean values, it is here accepted that this one
value is best attributed to a sampling error rather than to evidence for a local origin.
Summary
The Iva site investigations revealed a component that included Late Woodland and
Middle Mississippian ceramics, some in direct association with one another. The
grit-tempered Late Woodland sherds (n = 213) are dominated by Angelo Punctated,
the type affiliated with the Lewis phase, but also include a few Madison
Cord-Impressed sherds and one Aztalan Collared rim. The Mississippian pottery
(n = 90) includes rolled-rim varieties of Ramey Incised and Powell Plain; consequently, this component is another example of the Ramey horizon that is contemporaneous with the Stirling phase at Cahokia (A.D. 1100–1175). In addition, at
least two vessels with Powell Plain-like forms were cordmarked and grit tempered.
Thin sections from seven grit-tempered, cordmarked vessels and six shelltempered vessels were subjected to petrographic analysis in order to determine
their compositions for the purpose of distinguishing locally made from imported
vessels. Assuming the two Angelo Punctated vessels that were thin sectioned were
made of local materials, all but one of the other five grit-tempered vessels had
similar paste values, suggesting that six of them were made locally. The one exception, Madison Cord-Impressed vessel 47-336, has a siltier and sandier paste (see
Figure 11 and Table 3), suggesting a possible nonlocal derivation. Using body
values to distinguish local from nonlocal products is more problematic because
the amount and size of temper, the salient properties that distinguish bodies in
this case (all have granitic tempers), are more a reflection of pottery-making practices than of sources. It is interesting to note that among the four main stylistic
groups of grit-tempered vessels (Angelo Punctated, Madison Cord-Impressed,
Aztalan Collared, and grit-tempered Powell-like) there is an apparent clustering
into four groups that is suggestive of distinctive pottery-making practices for each.
The origin of Angelo Punctated remains unknown but seems to be related to an
eastward wave of influences from Great Oasis/Mill Creek peoples in southwestern
Minnesota and northwestern Iowa (Boszhardt 1996). For example, the Late
24
ROBERT F. BOSZHARDT AND JAMES B. STOLTMAN
Woodland pottery assemblage from Hartley Fort exhibits strong ties to Mill Creek
(Tiffany 1982, 1991), and isolated Great Oasis/Mill Creek sherds are also reported
from the Lundy and John Chapman sites at the Apple River locality of northwest
Illinois (Emerson et al. 2007:49–50; Millhouse 2012) and at the Cade 9 site in the
Bad Axe Valley (James Theler, personal communication 2009). Bennett (1945:
Plate 22f) illustrates a grit-tempered, angular-shoulder sherd with angular punctations and incised lines from the John Chapman Village along the Apple River that
may be related to Angelo Punctated, and Millhouse (2012:164) describes another
probable Angelo rim in his more recent synthesis of John Chapman.
Most recent summaries of Bennett phase components in the Apple River indicate
very close relations to the isolated Fred Edwards site in southwest Wisconsin (e.g.,
Emerson et al. 2007; Finney 2013; Millhouse 2012). Stoltman (1991b) has demonstrated through petrography that some pots found at Fred Edwards were manufactured at Hartley Fort, but he also views the Late Woodland assemblage at Fred
Edwards as an intrusion of Maples Mills potters from central Illinois (Finney and
Stoltman 1991; Stoltman and Christensen 2000). In 2008, a castellated Maples
Mills rim was recovered at the Lower Sand Lake site in an assemblage that includes
72 Angelo Punctated sherds, 11 Mississippian sherds, and 3 pseudo-collared rims
(Boszhardt and Holtz-Lieth 2008).
Given the apparent Plains influences of Angelo Punctated, it is possible that Lewis
phase peoples entered the northern portion of the Driftless Area from the west
around A.D. 950–1000. By the time early Mississippians established colonies at
the mouth of Coon Valley and Trempealeau in A.D. 1050, Angelo seems to have
been the predominant local Late Woodland ware in the Lewis phase territory.
Indeed, Angelo Punctated rim sherds were found with early Mississippian materials
in a clay-capped basin (Feature 1) at the Pelkey site in Trempealeau (Benden et al.
2011; Boszhardt et al. 2013). That is the only observed case of direct interaction
between Lewis phase peoples and Mississippians at the Fisher and Trempealeau
colonies. Fewer than 30 Angelo sherds have been recovered from the expansive
Fisher Mounds Site Complex, only one of which was found in the fill of a Mississippian feature (Boszhardt et al. 2010:11, 38). Another of the Angelo sherds from
Fisher is a unique collared rim (see Arzigian 2008; Benden 2004; Boszhardt et al.
2010; Stoltman et al. 2008).
The direct association of at least one variety of Angelo Punctated with Ramey
Incised at Iva reveals that this Lewis phase type persists after A.D. 1100. That
variety is distinctive in having wider lines and punctates and inner lip tool
impressions. A few comparable rims have been reported from the Silvernale phase
Diamond Bluff site in the Red Wing Locality (Hall 1962:133–134, Plate 41:R, S;
Rodell 1997:341). In addition, two angular shoulder sherds from Angelo Punctated
vessels at Iva indicate blending of this type with Middle Mississippian vessel form.
The presence of an Aztalan Collared rim at Iva is also telling of the Terminal Late
Woodland nature of this component. Aztalan Collared has been found across southern
Wisconsin including at Late Woodland–Middle Mississippian contact sites such as
Aztalan and Fred Edwards, as well as at a number of Driftless Area rockshelter sites
(e.g., Rosenbaum [Stoltman 1976], Mayland Cave [Storck 1972], Gottschall [Salzer
and Rajnovich 2000], and Viola [Boszhardt 1990:12]). Isolated occurrences of this
PETROGRAPHIC ANALYSIS OF LATE WOODLAND AND MIDDLE MISSISSIPPIAN CERAMICS
25
type or related collared forms are also reported from the sites along the Apple River
(Millhouse 2012). Aztalan Collared, Starved Rock Collared, and Point Sauble Collared
in eastern Wisconsin are dated to the period between ca. A.D. 1050 and A.D. 1150
(Kelly 2003). The Aztalan Collared rim from Iva may have been imported from southeastern Wisconsin along with a Baraboo pipestone ear spool.
The entire Iva assemblage includes only three Madison-ware sherds. The few
Madison Cord-Impressed sherds may represent Eastman phase vessels from the
southern Driftless Area or a pre-Angelo phase occupation in the northern Driftless
Area.
Most of the Mississippian vessels from Iva are shell tempered and represent
Powell Plain and Ramey Incised types. The petrographic analysis of six shelltempered sherds represents four Ramey Incised vessels, a bowl, and a bottle. Two
of the shell-tempered Ramey Incised vessels have compositional characteristics
that indicate local manufacture. However, the other two shell-tempered Ramey
Incised vessels and the bowl and bottle compare favorably with compositional attributes of previously analyzed samples from the American Bottom, and all have burnished, slipped surfaces. Consequently, the latter four vessels likely represent imports
from the Cahokian metropolitan area.
Discussion
The period from A.D. 1050 to A.D. 1200 was one of tremendous cultural transition
across much of the midcontinent that included both colonization and interaction as
reflected by ceramic vessels (and other objects) found great distances from their
place of manufacture, technological hybridization and innovation that was fostered
in multiethnic communities, and the establishment of palisaded villages as a response
to social tension. In the Driftless Area, this period marks the termination of the Late
Woodland Effigy Mound culture. Theler and Boszhardt (2006) have argued for a
population increase and landscape packing just before the end of the Late Woodland
in the Driftless Area. The predicted consequence of this scenario would have been
reduced mobility and overexploitation of key resources, necessitating subsistence
and settlement changes. As such, Terminal Late Woodland societies in that region
may have been ripe for abandoning not only large geographic areas but also longstanding religious traditions, such as effigy mound building, particularly if introduced to a “new way” via contact with early Mississippians.
That contact was initiated around A.D. 1050 through very early Middle Mississippian colonies that were founded at Stoddard and Trempealeau, toward the far
reaches of Cahokia’s northern hinterland (Pauketat et al. 2015). Neither of those
colonies was established at an extant Late Woodland village, but effigy mound
settlements and mound clusters were situated nearby. Indeed, the Fisher Mounds
site colony at Stoddard appears to have been purposefully located in a
no-man’s-land buffer zone between Eastman and Lewis phase effigy mound
peoples. Yet, there is also no evidence for local conflict during this period, with
both colonies being situated in nondefensive settings along the Mississippi River
proper and apparently lacking palisades. At Trempealeau, there is only a single
26
ROBERT F. BOSZHARDT AND JAMES B. STOLTMAN
example of direct contact between Lewis phase Late Woodlanders and Mississippian
colonists (Feature 1 at the Pelkey site).
The Fisher and Trempealeau colonies ended before the onset of the Ramey
horizon (ca. A.D. 1100). It is not known what became of these earliest Mississippian
colonists to the UMRV, but there is currently no clear evidence of a local or regional
transition to a subsequent archaeological expression. Instead, Ramey horizon manifestations in the Driftless Area are represented by components that contain mixed
assemblages of newer Woodland ceramic types, hybrid Woodland–Mississippian
forms, and sometimes Ramey Incised or Powell Plain vessels that were imported
from the American Bottom. Representative Ramey horizon components within
the Driftless Area include Fred Edwards, Hartley Fort, and Iva. Each of these occupations appears to have been relatively short-lived. Two much more substantial
Ramey horizon complexes developed at Apple River and Red Wing, both of
which appear to have transcended the beginning of the Oneota culture, between
ca. A.D. 1150 and A.D. 1250.
The Fred Edwards and Hartley Fort settlements sprang up in remote, interior settings that were within recently abandoned Effigy Mound territories. Both were palisaded and have non local ceramics that include Maples Mills from central Illinois
and Mill Creek from northwest Iowa and some Middle Mississippian materials
that were imported from the south, as well as hybrid vessels that indicate local blending of multiple cultural expressions. Local Woodland ceramics at Fred Edwards
include the Grant series (Finney 2013; Finney and Stoltman 1991), and Stoltman
(2000:444) interprets both Fred Edwards and Hartley Fort as site unit intrusions.
Iva is set back from the Mississippi River similar to Fred Edwards and Hartley
Fort, but unlike them, Iva does not appear to have been palisaded. Instead,
this location may have been selected for its rich agricultural soils at the mouth of
Sand Lake Coulee, as was clearly the case for its selection by subsequent Oneota
farmers.
The Red Wing and Apple River complexes developed on the north and south
margins of the Driftless Area, respectively, and each marks the corresponding
edge of effigy mound distributions. These mound and village clusters also appear
to have been populated by peoples from multiple regions, including effigy mound
peoples as they abandoned the Driftless Area. Theler and Boszhardt (2006)
suggested that Lewis phase peoples migrated to Red Wing, while former Eastman
phase peoples shifted to Apple River. Rodell (1991) and Millhouse (2012) have
argued that Mississippian influences came to established Late Woodland communities at Red Wing and Apple River, respectively. There are Woodland materials
and effigy mounds at both places, but the quantity at both localities is relatively
minor, particularly when compared to Fred Edwards and Iva. The village occupants
at Red Wing and Apple River also transformed their material culture and life ways
to become early Oneota manifestations, whereas those at Fred Edwards, Iva, and
presumably Hartley Fort did not. There are no early Oneota sites in the vicinity
of Fred Edwards, and those at Sand Lake Coulee and the La Crosse locality
appeared after A.D. 1300. There is an Oneota component at Hartley Fort, but it,
too, seems to correspond to a post-A.D. 1300 component.
PETROGRAPHIC ANALYSIS OF LATE WOODLAND AND MIDDLE MISSISSIPPIAN CERAMICS
27
At Red Wing, large Silvernale phase settlements reveal the transition from terminal Lewis phase influences to emergent Oneota with continued influence, if not
direct contact, from Ramey horizon Middle Mississippians (Boszhardt and Goetz
2000; Rodell 1991; Theler and Boszhardt 2003). Holley (2007) has reevaluated a
large sample of emergent Oneota ceramics from sites at Red Wing that has resulted
in a refined chronology that divides the Silvernale phase into early (A.D. 1150–
1200) and late (A.D. 1200–1250) segments (Emerson et al. 2007:53–54). Of relevance to the Lewis phase is the presence of a few Angelo Punctated sherds at the
Diamond Bluff site, which are similar to the larger vessel from Feature 16 at Iva,
and a “panther” mound that contained Mississippian and emergent Oneota
vessels (Hall 1962; Rodell 1991). All the Ramey pots at Red Wing seem to be
local emulations, but there are a few classic Middle Mississippian indicators, such
as a long-nose god masquette from Diamond Bluff.
A similar phenomenon appears to have occurred in the southern Driftless Area
with a depopulation of the former Eastman phase territory coinciding with the
Ramey horizon and rise of the Bennett phase at the Apple River complex. Stoltman’s
petrographic analysis of several Ramey vessel rims from the John Chapman site
found some to be probable imports from the American Bottom and others to be
local imitations (Millhouse 2012:150). Millhouse interpreted the Middle Mississippian platform mounds, ceramics, and lithics at Apple River as being due to a small
group of Mississippians arriving around A.D. 1100 in a locale long occupied by Late
Woodlanders. The initial response was to accept some of the new ideological concepts and materials, but in short order, the population became “creolized”,
melding into a local expression that ultimately became Oneota by A.D. 1250. He
agrees with Stoltman that the Mississippianization at nearby Fred Edwards likely
came through the larger Apple River complex.
Mechanisms for cultural interaction during the transitional period from A.D.
1050 to A.D. 1200 have been difficult to discern but appear to have included adoption of the Middle Mississippian worldview through ritual ceremonies that may
have included exchange of pottery, such as the Ramey Incised type (e.g., Pauketat
and Emerson 1991). Based on the co-occurrence of early variety Powell Plain and
Angelo Punctated in Feature 1 at the Pelkey site, it is clear that the pre-Ramey Mississippian colonists at Trempealeau made contact with Lewis phase peoples around
A.D. 1050. The Mississippian assemblages at Trempealeau do not include Ramey
Incised, but there is a high incidence of small, fine-ware bowls, many of which
were red slipped and nearly all of which were imported (Pauketat et al.
2015:278–282). These well-made red bowls would not have been suitable for
large feast events but likely impressed local people along with the other exotica
that the Mississippians brought along. Nonetheless, the local Late Woodlanders
do not appear to have adopted a Mississippian worldview at that time; instead
they maintained a degree of separation.
At Iva, unambiguous evidence of more involved cultural interaction between Late
Woodlanders and Mississippians during the Ramey horizon occurs in the form of
the co-occurrence of locally manufactured Late Woodland vessels, imported
Middle Mississippian vessels, and hybrids of the two traditions. Feature 16 at Iva
contained rims from at least two Angelo Punctated vessels in direct association
28
ROBERT F. BOSZHARDT AND JAMES B. STOLTMAN
with an imported Ramey (“weeping-eye”) vessel, another imported bowl, and
another apparent local emulation of a Ramey pot. That feature also contained the
remains of a sacrificial dog, high-quality deer cuts, corn, sunflower, wild rice,
tobacco, and other ecofacts, all of which suggest a feasting event, much like might
be envisioned in a power-of-the-pot ritual during the Ramey horizon. Clearly,
between A.D. 1100 and A.D. 1150, Lewis phase peoples were accepting and incorporating Middle Mississippian cultural aspects.
Of the five CCS defined by Stoltman (2000), Iva seems to conform to either
CCS 1 or CCS 3. Situation 1 is “identified through the presence of a limited
number of diagnostic, Cahokia-derived or inspired, portable artifacts (trait unit
intrusions…) within the context of an otherwise local cultural assemblage”
(Stoltman 2000:441–442). The Cahokian traits include Powell Plain and
Ramey Incised types or their imitations and can include marine-shell beads, longnosed god masquettes, pulley-shaped ear spools, and tri-notched points (Stoltman 2000:441–442). Situation 3 “involves the occurrence of a minority of
Cahokia-related traits (including both Powell Plain and Ramey Incised pottery)
within the context of a predominately Late Woodland assemblage…[T]he variants of the Late Woodland culture involved have no known local antecedents”
(Stoltman 2000:444–445). Scenario 3 is best exemplified by Hartley Fort and
Fred Edwards, both of which were interpreted as site unit intrusions (Stoltman
2000:444).
Two-thirds of the diagnostic sherds from Iva are Late Woodland, while nearly a
third are attributed to Mississippian potters or their influences. The local Late
Woodland assemblage is dominated by Angelo Punctated, which is the diagnostic
type of the Lewis phase. Based largely on the co-occurrence of Angelo Punctated
and early Mississippian materials at Pelkey, Lewis phase peoples must have been
established in the northern Driftless Area for at least a few decades prior to the
Ramey horizon Iva component. Because the early Mississippian colonies at Fisher
and Trempealeau were abandoned before A.D. 1100, Iva does not appear to have
been a direct offshoot from either of those. However, a few red-slipped, shelltempered sherds found during the surface collections at Iva (Boszhardt 2004:66)
may indicate an initial Mississippian presence at Sand Lake Coulee as Fisher and
Trempealeau were winding down.
Petrographic analysis indicates that imported Ramey Incised and Powell Plain
vessels at Iva were probably made in the American Bottom. These vessels may
have been brought to Sand Lake Coulee via exchange networks that connected
Bennett phase components at Apple River and Fred Edwards/Hartley Fort,
perhaps after having been transported through contemporary Mississippian settlements in the central Illinois River valley. Alternately, they may have been brought
directly to Sand Lake by Cahokians. At this time, it is simply not known if the
Lewis phase component existed at Sand Lake Coulee before the onset of Mississippian interaction (CCS 1) or if Iva represents a site unit intrusion of already blended
Woodland and Mississippian peoples (CCS 3). Whatever the case, the interaction
persisted long enough for local potters to adopt Mississippian recipes and styles
and to create hybrid vessels.
PETROGRAPHIC ANALYSIS OF LATE WOODLAND AND MIDDLE MISSISSIPPIAN CERAMICS
29
Conclusion
The Late Woodland–Middle Mississippian component at the Iva site represents
another example of direct interaction during the early portion of the Ramey
horizon (ca. A.D. 1100–1150). The Mississippian contact at Iva was with an established local Late Woodland entity—the Lewis phase, whose people occupied
the northern portion of the Driftless Area during the Terminal Late Woodland
period.
The Iva site, along with the early Mississippian component at Fisher Mounds and
the better-known complex at Trempealeau, provides opportunities to further
understand Middle Mississippian interaction from the American Bottom with
Late Woodland populations in the upper Midwest. Iva, in particular, appears to
have included a feasting event within the Lewis phase territory. Perhaps its setting
between the former American Bottom colonies at Fisher and Trempealeau indicates
an initial peaceful attempt at interaction by the “second wave” of Mississippians
into the northern hinterland. If so, the effort seems to have failed as no other substantial Mississippian sites are known in the La Crosse locality, while other contemporaneous sites (e.g., Fred Edwards and Hartley Fort) in the Driftless Area were
palisaded.
Acknowledgments
The authors thank developer Robert Thorpe for allowing repeated access to the Iva
site in 2002 and 2003, and to the field volunteers, most of whom were staff and
undergraduate students affiliated with from the Mississippi Valley Archaeology
Center at the University of Wisconsin-La Crosse. We also appreciate the efforts of
the MCJA editorial team and the productive comments of three reviewers, all of
which enhanced the content and clarity of this article.
Notes on contributors
Robert F. Boszhardt is an honorary fellow at the University of Wisconsin-Madison.
He received his MA in anthropology at the University of Wisconsin-Madison. His
research focus is the archaeology of the Upper Mississippi Valley with an emphasis
on the Driftless Area.
Correspondence to: Robert “Ernie” Boszhardt, 411 Sauk Street, Lodi, WI 53555,
USA. E-mail: rboszhardt@gmail.com.
James B. Stoltman is professor emeritus in the Department of Anthropology at the
University of Wisconsin-Madison. He earned his Ph.D. in anthropology at
Harvard University. His research has spanned much of North American archaeology
with an emphasis on the midcontinent. He specializes in ceramic petrographic
analyses.
Correspondence to: James Stoltman, Department of Anthropology, University of
Wisconsin-Madison, 1180 Observatory Drive, Madison, WI 53706, USA. E-mail:
Stoltman@wisc.edu.
30
ROBERT F. BOSZHARDT AND JAMES B. STOLTMAN
References Cited
Arzigian, Constance M. (2008) 2001–2006 Archaeological Studies at the Fisher Mounds Site (47Ve825),
Stoddard, Vernon County, Wisconsin. Reports of Investigations No. 635. Mississippi Valley Archaeology
Center, University of Wisconsin-La Crosse.
Bareis, Charles J., and James W. Porter, (editors) (1984) American Bottom Archaeology. University of Illinois
Press, Urbana.
Barrett, Samuel A. (1933) Ancient Aztalan, Vol. XII. Bulletin of the Public Museum of the City of Milwaukee,
Milwaukee, Wisconsin.
Benden, Danielle M. (2004) The Fisher Mounds Site Complex: Early Middle Mississippian Exploration in the
Upper Mississippi Valley. Minnesota Archaeologist 63:7–24.
Benden, Danielle M., Robert F. Boszhardt, and Timothy R. Pauketat (2011) The Mississippian Initiative: Year 2
at Trempealeau. The Wisconsin Archeologist 92:131–132.
Bennett, John W. (1945) Archaeological Explorations in Jo Daviess County, Illinois. University of Chicago
Press, Chicago.
Boszhardt, Robert F. (1985a) Oneota Ridged Fields at the Sand Lake Site (47Lc44) La Crosse County, Wisconsin
(with T. Bailey and J. Gallagher). The Wisconsin Archeologist 66:47–67.
Boszhardt, Robert F. (1985b) Final Cultural Resources Investigations along CTH “SN” in La Crosse County,
Wisconsin. Reports of Investigations No. 33, Mississippi Valley Archaeology Center, University of
Wisconsin-La Crosse.
Boszhardt, Robert F. (1989) Ceramic Analysis and Site Chronology of the Pammel Creek Site. The Wisconsin
Archeologist 70:41–94.
Boszhardt, Robert F. (1990) The Late Woodland Study Unit in Region 6, Western Wisconsin. Reports of
Investigations No. 115. Mississippi Valley Archaeology Center, University of Wisconsin-La Crosse.
Boszhardt, Robert F. (1996) Angelo Punctated: A Late Woodland Ceramic Type in Western Wisconsin. Journal
of the Iowa Archeological Society 43:129–138.
Boszhardt, Robert F. (1997) Ceramics from the Sand Lake Archaeological District, Wisconsin. Journal of the
Iowa Archeological Society 44:139–159.
Boszhardt, Robert F. (2004) The Late Woodland and Middle Mississippian Component at the Iva Site (47Lc42),
La Crosse County, Wisconsin. In the Driftless Area of the Upper Mississippi River Valley. Minnesota
Archaeologist 63:60–85.
Boszhardt, Robert F. (2009) Archaeological Data Recovery at the Meier Farm Development Within the Sand
Lake Archaeological District, La Crosse County, Wisconsin: 2002–2003. Reports of Investigations No.
576. Mississippi Valley Archaeology Center, University of Wisconsin-La Crosse.
Boszhardt, Robert F., Thomas W. Bailey, and James P. Gallagher (1985) Oneota Ridged Fields at the Sand Lake
Site (47Lc44) La Crosse County, Wisconsin. The Wisconsin Archeologist 66:1–46.
Boszhardt, Robert F., and Danielle M. Benden (2014) The Trempealeau Archaeology Project (TAP): The
Knepper Locality of the Uhl Site (47Tr-0159). In Archaeology Around Wisconsin. The Wisconsin
Archaeologist 95:136–138.
Boszhardt, Robert F., Danielle M. Benden, and Timothy R. Pauketat (2010) The Mississippian Initiative: Year 1
2009 Investigations at the Fisher Mounds Site Complex Stoddard, Wisconsin. Report prepared for the
National Science Foundation and St. Paul District Corps of Engineers.
Boszhardt, Robert F., Danielle M. Benden, and Timothy R. Pauketat (2012) The Mississippian Initiative: Year 3
(2011). The Wisconsin Archeologist 93:81–89.
Boszhardt, Robert F., Danielle M. Benden, and Timothy R. Pauketat (2013) The Pelkey Site (47Tr415) 2010
Discovery and Investigation: An Early Middle Mississippian Midden at Trempealeau, Wisconsin. Report submitted to the Wisconsin Historical Society.
Boszhardt, Robert F., Danielle M. Benden, and Timothy R. Pauketat (2015) Early Mississippian Outposts in the
North. In Medieval Mississippians: The Cahokian World, edited by Timothy R. Pauketat and Susan M. Alt,
pp. 63–69. School for Advanced Research Press, Santa Fe.
PETROGRAPHIC ANALYSIS OF LATE WOODLAND AND MIDDLE MISSISSIPPIAN CERAMICS
31
Boszhardt, Robert, and Natalie Goetz (2000) An Apparent Late Woodland Boundary in Western Wisconsin.
Midcontinental Journal of Archaeology 25:269–288.
Boszhardt, Robert F., and Wendy K. Holtz-Lieth (2008) Phase III Archaeological Investigations of a Portion of
47Lc45 for the Southern Half of the Proposed Elmwood Meier Farm East Development. Reports of
Investigation No. 762. Mississippi Valley Archaeology Center, University of Wisconsin-La Crosse.
Buckley, Ernest R. (1901) The Clays and Clay Industries of Wisconsin. Economic Series No. 4, Bulletin VII, Pt. I.
Wisconsin Geological and Natural History Survey, Madison.
Emerson, Thomas E. (1991) The Apple River Mississippian Culture of Northwestern Illinois. In Cahokia and the
Hinterlands: Middle Mississippian Cultures of the Midwest, edited by Thomas Emerson and R. Barry Lewis,
pp. 164–182. University of Illinois Press, Urbana.
Emerson, Thomas E., and Douglas K. Jackson (1984) The BBB Motor Site. American Bottom Archaeology
FAI-270 Site Reports 6. University of Illinois Press, Urbana.
Emerson, Thomas E., Dale L. McElrath, and Andrew C. Fortier (editors) (2000) Late Woodland Societies:
Tradition and Transformation Across the Midcontinent. University of Nebraska Press, Lincoln.
Emerson, Thomas E., Phillip G. Millhouse, and Marjorie B. Schroeder (2007) The Lundy Site and Mississippian
Presence in the Apple River Valley. The Wisconsin Archeologist 88:1–123.
Finley, Robert W. (1976) The Original Vegetation of Wisconsin. North Central Forest Experiment Station, U.S.
Department of Agriculture, St. Paul, MN.
Finney, Fred A. (1993) Hartley Fort Revisited: Late Woodland, Mill Creek, and Mississippian Culture Contact in
Northeast Iowa. Paper presented at the 37th Midwest Archaeological Conference, Grand Rapids, Michigan.
Copy on file, Office of the State Archaeologist, University of Iowa, Iowa City.
Finney, Fred A. (2013) Intrasite and Regional Perspectives on the Fred Edwards Site and the Stirling Horizon in
the Upper Mississippi Valley. The Wisconsin Archeologist 94:3–248.
Finney, Fred A., Russell P. Baldner, and E. Arthur Bettis III (1993) The 1993 Field Season at the Hartley Fort Site
in Northeast Iowa. Paper presented at the 38th Midwest Archaeological Conference, Milwaukee, Wisconsin.
Copy on file, Office of the State Archaeologist, University of Iowa, Iowa City.
Finney, Fred A., and James B. Stoltman (1991) The Fred Edwards Site: A Case of Stirling Phase Contact in
Southwestern Wisconsin. In New Perspectives on Cahokia: Views from the Periphery, edited by James B.
Stoltman, pp. 229–252. Monographs in World Archaeology No. 2. Prehistory Press, Madison, WI.
Flock, M. A. (1983) The Late Wisconsinan Savanna Terrace in Tributaries to the Upper Mississippi River.
Quaternary Research 20:165–176.
Frolking, T. A. (1982) The Genesis and Distribution of Upland Red Clays in Wisconsin’s Driftless Area. In Field
Trip Guidebook, No. 5, pp. 88–97. Wisconsin Geological and Natural History Survey, Madison.
Gallagher, James P. (1980) La Crosse Area Archaeological Survey, 1979 Season. Report prepared for the State
Historical Society of Wisconsin. Copy on file, Mississippi Valley Archaeology Center, University of
Wisconsin-La Crosse.
Gallagher, James P., Robert F. Boszhardt, Robert F. Sasso, and Katherine P. Stevenson (1985) Oneota Ridged
Field Agriculture in Southwestern Wisconsin. American Antiquity 50:605–612.
Gallagher, James P., Robert F. Boszhardt, Robert F. Sasso, and Katherine P. Stevenson (1987) Floodplain
Agriculture in the Driftless Area: A Reply to Overstreet. American Antiquity 52:398–404.
Gibbon, Guy E. (1979) The Mississippian Occupation of the Red Wing Area. Minnesota Prehistoric
Archaeology Series, No. 13. Minnesota Historical Society, St. Paul.
Gibbon, Guy E., and Clark A. Dobbs (1991) The Mississippian Presence in the Red Wing Area. In New
Perspectives on Cahokia: Views from the Periphery, edited by James B. Stoltman, pp. 281–306.
Monographs in World Archaeology No. 2. Prehistory Press, Madison, WI.
Green, William (1997) Middle Mississippian Peoples. In Wisconsin Archaeology, edited by Robert
A. Birmingham, Carol I. Mason and James B. Stoltman. The Wisconsin Archeologist 78:202–222.
Green, William (2015) Identity, Ideology, and the Effigy Mound-Oneota Transformation. The Wisconsin
Archeologist 95:44–72.
Green, William, and Roland Rodell (1994) The Mississippian Presence and Cahokia Interaction at Trempealeau,
Wisconsin. American Antiquity 59:334–359.
32
ROBERT F. BOSZHARDT AND JAMES B. STOLTMAN
Hall, Robert L. (1962) The Archaeology of Carcajou Point: With an Interpretation of the Development of
Oneota Culture in Wisconsin. University of Wisconsin Press, Madison.
Hall, Robert L. (1991) Cahokia Identity and Interaction Models of Cahokia Mississippian. In Cahokia and the
Hinterlands: Middle Mississippian Cultures of the Midwest, edited by Thomas E. Emerson and R. Barry
Lewis, pp. 3–34. University of Illinois Press, Springfield.
Holley, George R. (2007) The Red Wing Locality Late Prehistoric Ceramic Sequence. Manuscript on file,
Department of Anthropology and Earth Science, Minnesota State University, Moorhead.
Jackson, Douglas K., Andrew C. Fortier, and Joyce A. Williams (1992) The Sponemann Site 2: The Mississippian
and Oneota Occupations (11-Ms-517). University of Illinois Press, Urbana.
Kelly, John E. (1990) Range Site Community Patterns and the Mississippian Emergence. In The Mississippian
Emergence, edited by B. Smith, pp. 67–112. Smithsonian Institution Press, Washington, DC.
Kelly, John E. (1991a) Cahokia and Its Role as a Gateway Center in Interregional Exchange. In Cahokia and the
Hinterlands: Middle Mississippian Cultures of the Midwest, edited by Thomas E. Emerson and R. Barry
Lewis, pp. 61–82. University of Illinois Press, Springfield.
Kelly, John E. (1991b) The Evidence for Prehistoric Exchange and Its Implications for the Development of
Cahokia. In New Perspectives on Cahokia: Views from the Periphery, edited by James B. Stoltman, pp.
65–92. Monographs in World Prehistory, No. 2. Prehistory Press, Madison, WI.
Kelly, John M. (2003) Delineating the Spatial and Temporal Boundaries of Late Woodland Collared Wares from
Wisconsin and Illinois. Unpublished master’s thesis, Department of Anthropology, University of
Wisconsin-Milwaukee.
Ketterhagen, Lynn (2006) Prehistoric Clay Sources: A Forensic Study in Geoarchaeology. University of
Wisconsin-La Crosse. Journal of Undergraduate Research IX:1–5. Electronic document, http://www.uwlax
.edu/urc/JUR-online/PDF/2006/ketterhagen.pdf, Accessed: October 20, 2015.
Knox, James C. (1996) Late Quaternary Upper Mississippi River Alluvial Episodes and Their Significance to the
Lower Mississippi River System. Engineering Geology 45:263–285.
LeMaitre, R. W. (editor) (2002) Igneous Rocks, A Classification and Glossary of Terms. Cambridge University
Press, Cambridge.
Martin, Lawrence (1965) The Physical Geography of Wisconsin. University of Wisconsin Press, Madison.
McKusick, Marshal B. (1964) Discovering the Hartley Fort. Palimpsest 45:487–494.
Mickelson, David M., James C. Knox, and Lee Clayton (1981) Glaciation of the Driftless Area: An Evaluation of
the Evidence. In Quaternary History of the Driftless Area, edited by James C. Knox, Lee Clayton, and David
M. Mickelson, pp. 155–170. Field Trip Guide Book No. 5, Geological and Natural History Survey, University
of Wisconsin Extension, Madison.
Millhouse, Phillip G. (2012) The John Chapman Site and Creolization on the Northern Frontier of the
Mississippian World. Unpublished Ph.D. dissertation, Department of Anthropology, University of Illinois
at Urbana-Champaign. Electronic document, https://www.ideals.illinois.edu/handle/2142/31164, Accessed:
September 15, 2015.
Milner, George R. (1991) American Bottom Mississippian Cultures: Internal Developments and External
Relations. In New Perspectives on Cahokia: Views from the Periphery, edited by James B. Stoltman, pp.
29–48. Monographs in World Prehistory, No. 2. Prehistory Press, Madison, WI.
Overstreet, David F. (2000) Cultural Dynamics of the Late Prehistoric Period in Southern Wisconsin. In Mounds,
Modoc, and Mesoamerica: Papers in Honor of Melvin L. Fowler, edited by Steven R. Ahler, pp. 405–438.
Scientific Papers Series Vol. XXVIII, Illinois State Museum, Springfield.
Pauketat, Timothy R. (1984) Notes on Some Non-Local Ceramics from an American Bottom Collection. Illinois
Antiquity 61(1):4–6.
Pauketat, Timothy R. (1998) Refiguring the Archaeology of Greater Cahokia. Journal of Archaeological
Research 6:45–89.
Pauketat, Timothy R. (2004) Ancient Cahokia and the Mississippians. Cambridge University Press, Cambridge.
Pauketat, Timothy R. (2009) Cahokia: Ancient America’s Great City on the Mississippi. Viking-Penguin,
New York.
PETROGRAPHIC ANALYSIS OF LATE WOODLAND AND MIDDLE MISSISSIPPIAN CERAMICS
33
Pauketat, Timothy R., Robert F. Boszhardt, and Danielle M. Benden (2015) Trempealeau Entanglements: An
Ancient Colony’s Causes and Effects. American Antiquity 80:260–289.
Pauketat, Timothy R., and Thomas E. Emerson (1991) The Ideology of Authority and the Power of the Pot.
American Anthropologist 93:919–941.
Richards, John D. (1992) Ceramics and Culture at Aztalan: A Late Prehistoric Village in Southeast Wisconsin.
Unpublished Ph.D. dissertation, Department of Anthropology, University of Wisconsin-Milwaukee.
Richards, John D. (2007) Context and Process: Red-Slipped Pottery in Cahokia’s Northern Hinterlands. Illinois
Archaeology 19:1–26.
Rodell, Roland (1989) The Middle Mississippian Study Unit: Region 6, Western Wisconsin. Report of
Investigation No. 94. Mississippi Valley Archaeology Center, University of Wisconsin-La Crosse.
Rodell, Roland (1991) The Diamond Bluff Complex and Cahokia Influence in the Red Wing Locality. In New
Perspectives on Cahokia: Views from the Periphery, edited by James Stoltman, pp. 253–280. Monographs in
World Archaeology No. 2. Prehistory Press, Madison, WI.
Rodell, Roland (1997) The Diamond Bluff Complex: Time and Tradition in the Northern Mississippi Valley.
Unpublished Ph.D. dissertation, Department of Anthropology, University of Wisconsin-Milwaukee.
Rosebrough, Amy (2010) Every Family a Nation: A Deconstruction and Reconstruction of the Effigy Mound
“Culture” of the Western Great Lakes of North America. Unpublished Ph.D. dissertation, Department of
Anthropology, University of Wisconsin-Madison.
Salzer, Robert J., and Grace Rajnovich (2000) The Gottschall Rockshelter: An Archaeological Mystery. Prairie
Smoke, St. Paul.
Sasso, Robert F., Robert F. Boszhardt, James C. Knox, James L. Theler, Katherine P. Stevenson, James P.
Gallagher, and Cynthia Stiles-Hanson (1985) Prehistoric Ridged Field Agriculture in the Upper Mississippi
Valley. Reports of Investigations No. 38. Mississippi Valley Archaeology Center, University of
Wisconsin-La Crosse.
Stevenson, Katherine P., and Robert F. Boszhardt (1983) National Register of Historic Places nomination form
(NPS 10–900-a) for the Sand Lake Archaeological District. Copy on file Mississippi Valley Archaeology
Center, University of Wisconsin-La Crosse.
Stoltman, James B. (1976) Two New Late Woodland Radiocarbon Dates from the Rosenbaum Rockshelter (47
Da411) and Their Implications for Interpretations of Wisconsin Prehistory. The Wisconsin Archeologist
57:12–28.
Stoltman, James B. (1989) A Quantitative Approach to the Petrographic Analysis of Ceramic Thin Sections.
American Antiquity 54:147–160.
Stoltman, James B. (1991a) Cahokia as Seen from the Peripheries. In New Perspectives on Cahokia: Views from
the Periphery, edited by James B. Stoltman, pp. 349–354. Monographs in World Archaeology No. 2.
Prehistory Press, Madison, WI.
Stoltman, James B. (1991b) Ceramic Petrography as a Technique for Documenting Cultural Interaction: An
Example from the Upper Mississippi Valley. American Antiquity 56:103–120.
Stoltman, James B. (2000) A Reconsideration of the Cultural Processes Linking Cahokia to Its Northern
Hinterlands During the Period A.D. 1000–1200. In Mounds, Modoc, and Mesoamerica: Papers in Honor
of Melvin L. Fowler, edited by Steven Ahler, pp. 439–454. Illinois State Museum Scientific Papers, Vol.
XXVIII, Springfield.
Stoltman, James B. (2001) The Role of Petrography in the Study of Archaeological Ceramics. In Earth Sciences
and Archaeology, edited by Paul Goldberg, Vance Holliday, and Reid Ferring, pp. 297–326. Kluwer
Academic/Plenum, New York.
Stoltman, James B., Danielle M. Benden, and Robert F. Boszhardt (2008) Evidence in the Upper Mississippi
Valley for Pre-Mississippian Cultural Interaction with the American Bottom. American Antiquity
73:317–336.
Stoltman, James B., and George W. Christensen (2000) The Late Woodland Stage in the Driftless Area of the
Upper Mississippi Valley. In Late Woodland Societies: Tradition and Transformation Across the
Midcontinent, edited by Thomas E. Emerson, Dale L. McElrath, and Andrew C. Fortier, pp. 497–524.
University of Nebraska Press, Lincoln.
34
ROBERT F. BOSZHARDT AND JAMES B. STOLTMAN
Storck, Peter L. (1972) The Archaeology of Mayland Cave. Unpublished Ph. D. dissertation, Department of
Anthropology, University of Wisconsin-Madison.
Theler, James L., and Robert F. Boszhardt (2003) Twelve Millennia: Archaeology of the Upper Mississippi
Valley. University of Iowa Press, Iowa City.
Theler, James L., and Robert F. Boszhardt (2006) Collapse of Crucial Resources and Culture Change: A Model
for the Woodland to Oneota Transformation in the Upper Midwest. American Antiquity 71:433–472.
Tiffany, Joseph A. (1982) Hartley Fort Ceramics. Proceedings of the Iowa Academy of Science 89:133–150.
Tiffany, Joseph A. (1991) Models of Mississippian Culture History in the Western Prairie Peninsula: A
Perspective from Iowa. In Cahokia and the Hinterlands: Middle Mississippian Cultures of the Midwest,
edited by Thomas Emerson and R. Barry Lewis, pp. 183–192. University of Illinois Press, Urbana.
View publication stats