New Investigations of the Cates Farm Site, Vassalboro, Maine
Gary D. Shaffer, Arthur E. Spiess, and Nancy Asch Sidell
Natural Resources Conservation Service; Maine Historic Preservation Commission;
Archeobotanical Consulting
Introduction
The Cates Farm archaeological site (prehistoric
site 38.10) is located at the outlet of China
Lake, in East Vassalboro. Artifacts collected
from the site cover much of the span of Maine’s
pre-Eur-opean occupation, from Early Archaic
through Late Ceramic (Trautman and Spiess
1992; Trautman 1996). Occupations by various
Middle and Late Archaic cultures appear to
have left most of the artifacts.
Reconstructions of Archaic period subsistence and settlement patterns have become
possible with the accumulation of relevant data
(faunal remains and charcoal) in recent years
(e.g., Asch Sidell 1999; Spiess and Mosher
2006), necessarily focused on interior river and
lake-side sites because of the effects of sea level
rise in eroding coastal sites (see Kelley et al.
2010). Harvesting of diadromous and fresh
water fish (such as sturgeon, alewife, eel, and
sucker) were a large part of Archaic subsistence
(Spiess and Halliwell 2011), and harvesting
seasonal “runs” of anadromous fish must have
had an effect on settlement patterns. People
camped at the right time and place to harvest the
fish runs, and constructed and maintained fish
weirs such as the one in Sebasticook Lake
(Petersen et al. 1994).
The Natural Resources Conservation Service (NRCS) archaeological work at the Cates

Farm reported here adds modest but important
information that alewife were harvested at the
site 4000 years ago. This is not an unexpected
piece of news. But restoration of anadromous
fish in the lower Kennebec River, river herring
(including alewife) included, has been a focus
of effort since removal of Edwards and Fort
Halifax dams. In particular, restoration of the
alewife run up Outlet Stream into China Lake is
a current focus of multiple fisheries and related
agencies, including NRCS, and non-profit
groups (McGuire 2015). Restoration of alewife
runs has been a state-wide priority for various
groups because of the perceived benefits to Gulf
of Maine fisheries, among other things (Watts
2012), but not without controversy. In fact, the
identification of alewife bones in Archaic and
Ceramic period sites on the Spednic Lakes of
the Saint Croix drainage was cited in the Maine
legislature in arguments over restoring alewife
runs to the Saint Croix (e.g., Miller 2015; Arthur Spiess to Paul Bisulca, personal communication April 29, 2015). Thus, the information
reported herein is confirmation of the antiquity
of the alewife run into China Lake. Documentation of alewife harvesting for at least 5000
years at the outlet of China Lake provides
support for NRCS’s planning to restore alewife
runs in the Outlet Stream by creating fish passage through several dams.

The Maine Archaeological Society Bulletin 55(2):1-15 (2015)

 New Investigations of the Cates Farm Site, Vassalboro, Maine

Cates Farm Site
Archaeological Background

encountered three features. The entire site area
is 300 x 150 m, with the most intensely occupied area near the outlet being about 100 m x
100 m. Extrapolating from the 1990 and 2014
testing, it is possible that there might be hundreds of features below the plowzone.
Stone artifacts in the Cates Farm collection,
mostly now with the Vassalboro Historical
Society, include multiple pecked and ground
stone tools: adzes, axes, gouges, celts, stone
rods, plummets, and at least two ground slate
points (Trautman and Spiess 1992; Trautman
1996). These objects were described and reported prior to Petersen’s (1991) publication of
descriptions of well-dated Archaic ground and
pecked stone from the Sharrow site. In retrospect, the Cates Farm collection’s ground and
pecked stone appears to cover the Middle and
Late Archaic. Diagnostic stone points from the
Cates Farm include one Early Archaic Kirk-like
point, two Neville-like points, four Otter Creek
points, other Laurentian Archaic points, and at
least three Susquehanna Broad points (Trautman and Spiess 1992:17-34). Diagnostic Ceramic period points are primarily Early Ceramic
in age. Because of the plowed deposits and
limited sub-plowzone feature recovery, we do
not have enough information to determine the
presence or absence of the “Gulf of Maine Archaic” (which is mostly without diagnostic
bifaces).
There are few sites known around China
Lake or along its Outlet Stream, perhaps because of a lack of systematic archaeological

The site at the mouth of China Lake on the east
side of Outlet Stream in East Vassalboro was
initially assigned site number 38.10 by Dean
Snow, in the 1969 initial compilation of Maine
prehistoric sites. Bruce Bourque visited the site
in 1976, and observed in his 5-page “Report on
an Archaeological Survey of the Lower Kennebec Valley and Adjacent Areas” (Bourque
1976) that a “recheck of 38.10 …indicated
extensive plowing for generations. Artifacts
recovered during this period are in the hands of
family members and in the Vassalboro Historical Society.” At the request of the landowners,
Maine Historic Preservation Commission
(MHPC) did archaeological testing of the site in
1990. MHPC produced a topographic map of
the site and excavated seven 1x1 m test units
and three 50 x 50 test pits, recovering 51 tools,
217 pieces of debitage, and 220 calcined faunal
bones (2 turtle, 5 mammal, 10 small fish including one Alosine, the rest unidentified). The
details, as well as reporting of the Vassalboro
Historical Society collection and family collection are presented in the previously-cited Trautman and Spiess (1992) and Trautman (1996).
Excavation of almost eight square meters by
MHPC revealed two pre-European probable
garbage pits (Features 2 and 5). The 2014
testing reported herein used three 50 x 50 cm
test pits, and encountered a third sub-plowzone
feature. Thus, 8 ½ square meters of testing
2

 New Investigations of the Cates Farm Site, Vassalboro, Maine

survey. However, there are also few avocational archaeologists’ reports, so the raised water
levels in China Lake (outlet dam) and other
dams along Outlet Stream might be affecting
site visibility by inundation. A couple of linear
rock features that might be fish weirs have been
reported in Outlet Stream (site 53.95, Doug
Watts, personal communication 2014). The
situation changes as Outlet Stream joins the
south bank of the Sebasticook River, one kilometer upstream from the Sebasticook’s confluence with the Kennebec. Within one kilometer
of the Sebasticook/Outlet Stream confluence,
downstream to the mouth of the Sebasticook at
Fort Halifax, there are 20 prehistoric site numbers assigned. These include mostly eroded or
partially eroded sites with Middle Archaic,
Otter Creek, Moorehead phase, Susquehanna
tradition and Ceramic period components
(MHPC site files). At the Sebasticook-Kennebec confluence, under Fort Halifax, a stratified
prehistoric site begins with a circa 3100-yearold occupation, overlain by multiple Ceramic
period levels (site 53.35; Spiess, unpublished).
Fish bone in that site includes salmon and sturgeon (Spiess and Halliwell 2011:Figure 3).
Multiple sites on the Kennebec downstream
from the confluence have been tested, collected
and published in some cases (e.g., Spiess 1999;
Spiess and Hedden 1999, 2000), including much
material found when Edwards Dam was removed and the water levels returned to natural
flow (Spiess 2003, 2004; Spiess and Cranmer
2000).

Testing and Analytical Studies
in 2014 and 2015
In June 2014 Shaffer conducted limited archaeological testing of site 38.10 where NRCS
planned to assist the Cates family with installation of a seasonal high tunnel (hoop house).
Background research had indicated that site
38.10 has been the scene of avocational collecting by the family, as well as of professional
archaeological survey by MHPC (Trautman
1996; Trautman and Spiess 1992). That work,
as mentioned above, recovered artifacts primarily from the Middle and Late Archaic periods,
while the Early Archaic, Ceramic, and historic
periods were also represented. Most intensive
occupation seemed to be during the Vergennes
Phase (ca. 6000 – 4500 B.P.). The MHPC has
found the site to be eligible for the National
Register of Historic Places.
Installation of the high tunnel on the cultivated land would entail no grading or other site
preparation. The only ground disturbance
associated with its installation or use would be
minimal: the setting of stakes to anchor the
structure to the ground. Still, as a federal agency conducting a review under Section 106 of the
National Historic Preservation Act, NRCS’s
assessment of adverse effects includes consideration of “reasonably foreseeable effects caused
by the undertaking that may occur later in time
…” (36 CFR 800.5.a.1). In that light, one
should take into account that a high tunnel, in

3

 New Investigations of the Cates Farm Site, Vassalboro, Maine

Figure 1. View to the north of the cultivated ground proposed for the high tunnel and tested
with a walkover and shovel testing.
client Chris Cates laid out the corners of the
area (ca. 50 m east of China Lake’s outlet
stream) and noted that most artifacts had been
found in previous years in the western part of
this lower garden rather than in this eastern
section (see Trautman and Spiess 1992: 40).
The soil had been harrowed and was bare of
vegetation, providing excellent visibility of any
artifacts (Figure 1). The survey crew walked in
parallel transects separated by ca. 2 m and
found four sherds of historic ceramics. These

the future, might include the addition of appur
tenances like buried irrigation pipes that could
impact archaeological resources. Accordingly,
it seemed warranted to examine that part of site
38.10 where the high tunnel would be set so as
to provide information for any historic preservation conditions needed for installation of the
farm structure.
NRCS’s archaeological testing began with
a walkover survey of the 30-x-72-ft (9-x-22- m)
footprint of the planned high tunnel. NRCS
4

 New Investigations of the Cates Farm Site, Vassalboro, Maine

Table 1. Artifacts Recovered during the Testing of Site 38.10.
Provenience
Surface of proposed footprint of high tunnel

STP 5 m/Ap/0-33 cm Below Surface (B.S.)

STP 10 m/Ap/0-33 cm B.S.

STP 10m/Feature fill/43-53
cm B.S.
STP 15 m/Ap/0-28 cm B.S.

Artifact/Material
1 body sherd salt-glazed
stoneware
2 rim sherds blue-edged
whiteware
1 basal sherd bone china
1 nail or wire fragment
3 pieces coal
1 body sherd whiteware
1 body sherd redware
1 quartz fragment (flake?)
1 Kineo rhyolite flake
1 piece coal
1 rim sherd blue-edged
whiteware
1 body sherd whiteware
1 body sherd glazed
redware
1 iron spike
Calcined bone, charcoal
1 iron spike
1 United States Winged
Liberty Head (“Mercury”)
dime

Comments
Stoneware 1705-1930, blueedged whiteware c. 18301860, bone china 1794present (Brown 1982; Miller 2000)
Whiteware 1820 – present,
redware 19th century
(Brown 1982; Miller 2000);
rhyolite flake (medial – distal fragment) weighs 5.8 g
Spike 10.5 cm long

See identifications by
Spiess and Asch Sidell
Spike 9.8 cm long; dime
minted 1919

Next, Shaffer excavated three 50-x-50-cm
shovel test pits (STPs) to examine the length
of the high tunnel area. He measured in these
pits down the centerline of the high tunnel,
with the first at 5 m south of the northern end
and the other two at 10 m and 15 m south of
the northern end. Accordingly, the test pits

items could date from the eighteenth through
twentieth centuries (Table 1). They represent
items discarded either by the Cates, who had
resided at the farm since about 1830 (Cates
family, personal communication 2008), or
earlier occupants.

5

 New Investigations of the Cates Farm Site, Vassalboro, Maine

were designated “STP 5m,” “STP 10m,” and
“STP 15m.” The spacing of the excavation
pits would provide sufficient areal coverage of
the tunnel’s footprint. The three NRCS test
pits are located just east of the “lower garden”
as mapped by Trautman and Spiess (1992:Figures 4 and 6). The NRCS test pits are approximately 30 m east of the 1990 N260 E100 and
N280E100 test units that located Features 2
and 5, respectively. STP excavation by hand
shovel proceeded by soil horizons and, for the
subsoil, by 10-cm levels within them. At the
base of the plowzone, the floor of the test pit
was troweled and examined for cultural features. Digging continued at least 10 cm into
the subsoil till artifact-free soils were encountered. Excavated soil was sifted through ¼inch mesh hardware cloth. Artifacts were collected according to STP and level. Recordation of the STPs involved compiling maps
with global positioning system points and triangulation to nearby farm buildings, drawing
STP soil profiles, noting cultural finds, and
taking appropriate photographs. NRCS will
curate these records and other survey notes in
its State Office in Bangor, Maine. Processing
of identified artifacts from the test pits and
surface collection entailed cleaning prior to
analysis and storage in acid-free polyethylene
bags labeled with provenience. The landowner will retain the artifacts.
Excavation of the STPs revealed Ap horizons ranging from 28 to 33 cm deep. These
plowzone soils were of 10YR3/2 (very dark

grayish brown) very fine sandy loam. The
underlying B horizons were also of very fine
sandy loam, while colors ranged from
10YR4/4 (dark yellowish brown) to 2.5Y5/3
(light olive brown). Test pit excavation continued down to depths generally of 42 or 43
cm below surface. These soils are consistent
with Scio very fine sandy loam, the sediment
mapped for the project area in the Kennebec
County soil survey (Faust and LaFlamme
1978). Scio very fine sandy loam soils are
found on sloping or gently sloping terraces
adjacent to streams and natural drainageways.
The soil can be wet in the spring and during
periods of heavy precipitation. The soils are
free of stones, and nearly all have been cultivated at one time, but the Scio soils are used
mainly as woodland and for hay and pasture
(Faust and LaFlamme 1978).
The Ap horizons of the three STPs yielded several historic artifacts and a smaller
quantity of prehistoric material. Historic
items included ceramic sherds dating from the
nineteenth to twentieth centuries, a United
States “Mercury” dime of 1919, coal, a nail or
wire fragment, and iron spikes. With respect
to the prehistoric period, excavations found
one flake of Kineo rhyolite and a chip (flake?)
of quartz (Table 1). No artifacts were recovered from sifting the B horizon sediment.
Importantly, troweling of the top of the B
horizon soils revealed a feature (later designated as Feature 2014) in STP 10m. Feature

6

 New Investigations of the Cates Farm Site, Vassalboro, Maine

2014 first appeared at 33 cm below surface
as an indistinct darker area. Further excava
tion of the B horizon to 43 cm below surface
showed this stain to become more defined
along the western wall of the STP (Figure 2).
A relatively thin rock sat on edge to the immediate east of the stain. The position of
this rock may indicate it served as a form of
site “furniture.” Earlier excavations by
Trautman and Spiess (1992:11) elsewhere at
the site found that the base of a garbage pit
(Feature 2, dated to 5000 ± 70 BP) sometimes rested on clusters of burned rock slabs.
Following photography and scale-drawing of
the newly identified feature, Shaffer excavated its dark soil and bagged it for more
detailed analysis. The feature’s fill consisted
of 10YR2/1 (black) silt loam with a greasy
feel. This dark fill continued down to a
depth of 53 cm below surface, making for a
shallow pit at least on this edge of the feature
(Figure 3). It is possible the feature is deeper
to the west. A profile view of Feature 2014
reveals that some intact B horizon soil covers
the feature (Figure 4).
Processing of Feature 2014 began with
flotation of 3.1 kg of moist fill in a clean tub
with water, agitation of the fill with a plastic
rod, and recovery of a light fraction from the
water’s surface with a clean brass testing
sieve (no. 60 mesh or 250 micrometers).
The non-floating heavy fraction was placed
on 1 mm mesh hardware cloth and waterscreened. Following drying, the fractions

Figure 2. STP 10m (50 x 50 cm) with the surface (in
the B horizon) at 43 cm below surface and the dark
feature along the west (left) wall.

Figure 3. STP 10m with feature fill removed from
the floor (in the B horizon) at 43 cm to 53 cm below
surface (north at top). Note standing rock adjacent
to feature.

7

 New Investigations of the Cates Farm Site, Vassalboro, Maine

paratus from an alewife (river
herring; Alosa pseudoharengus).
Adult anadromous alewives
swim up streams in early May to
early June and spawn in lakes
and ponds. Surviving adults
swim downstream to the sea afterward; and young alewives
head seaward from July to November. Fish-eating birds and a
range of other fish and mammals
prey upon alewives. Human
consumption of alewives taken
Figure 4. STP 10m’s western wall profile following removal from Maine’s river fisheries deof the feature’s fill.
clined relative to other fish species in the twentieth century;
commercial demand for the fish
is high, however, as lobster bait
(Maine Department of Marine
Resources 2012).
The bone from the feature
fill in NRCS STP 10m is a calcined (burned) fragment of the
left angular bone (part of the jaw
apparatus) of an alewife (Figure
Figure 5. The calcined alewife bone (top) and a modern 5). Bone does not survive for
long after burial in Maine’s acid,
comparative specimen (cm scale).
interior soils unless it has been
calcined. Moreover, alewife bone is generally
were examined for cultural materials. No artiquite delicate, and few pieces of the skeleton
facts were identified, but biological items
were present: one or two calcined bone frag- would be expected to survive in archaeologiments and several pieces of charcoal.
cal context (see Spiess and Halli well 2011:5Spiess analyzed the calcined bone and 7 for calcination etc., and Figure 3 for images
succeeded in identifying one as the mouth ap- of delicate alewife bone). Only a few skeletal
8

 New Investigations of the Cates Farm Site, Vassalboro, Maine

elements of alewife seem to be “hard” enough
to survive to be recognizable in calcined archaeological assemblages: some vertebral
centra (backbone central parts), the round portion of the pro-otic bulla (ear bone at base of
skull), and limited portions of some of the jaw
apparatus. The hook-shaped portion of the
angular bone found in the NRCS STP 10 feature fill is one of these. The angular bone is
paired, and this one happens to be a left-side
specimen roughly 2 x 4 mm in size. Calcination shrinks bone by 10 to 30%, accounting
for the slightly smaller size of the archaeological specimen (Figure 5).
There are three closely related Alosine
fish species that would yield a similar bone
(Spiess and Halliwell 2011): alewife, blueback
herring (Alosa aestivalis), and shad (Alosa
sapidissima). The shad is noticeably larger
(about 2x linear dimension) than the other two
species. Mixed schools of river herring (blueback and alewife) begin their spawning migrations on Maine’s lower rivers and streams, but
the blueback herring (usually, 10% of the beginning of the run) drop out and do not proceed as far upstream as the inland lakes where
the alewife spawns. Thus, the Feature 2014
bone is identified as alewife or blueback herring based on size and morphology, but it
must be alewife based on inland location.
Nancy Asch Sidell analyzed the carbonized plant remains recovered by tub flotation
of the Feature 2014 fill. The sample was very
small--consisting of only 0.137 g of “dirty”

charcoal fragments larger than 0.5 mm. Visual
inspection (rather than sieving the fragile
specimens) determined that only ten wood and
one possible bark fragment were larger than 2
mm (Table 2). Uncarbonized plant remains
were assumed to be more recent inclusions
and were not tabulated. For the ten tiny wood
charcoal fragments in Feature 2014, the transverse section was first studied at 30X magnification after manually breaking the charcoal to
obtain a clean section. The red oak group
specimens were easily identified at that magnification. The cherry specimens were also
viewed on a tangential section at 400X to confirm the presence of diagnostic irregularlyspaced spiral thickenings on the vessels. The
white pine radial section viewed at 400X had
characteristic window-like cross-field pitting
and lacked dentate ray tracheids, which distinguishes it from red pine and pitch pine.
Table 2 also includes wood identifications
from Features 2 and 5, which were analyzed
in 1990. In those features, 20 fragments of
wood were identified from each sample using
a magnification of 30-70X. The larger sample
size in Features 2 and 5 revealed an even
greater number of wood types, including
beech, birch, pine, red oak group, ash, and
sugar maple. Red oak group includes only
northern red oak (Quercus rubra) in Kennebec
County (Maine Forest Service 2008).

9

 New Investigations of the Cates Farm Site, Vassalboro, Maine

Table 2. Carbonized Plant Remains from the Cates Farm Site.
Feature
Square

2

5

2014

N260E100

N280E100
Late
Archaic

STP 10m

C-14 date (BP)/Affiliation
5000+/-70
SAMPLE WEIGHT (g)
>2 mm
5
0.5-2 mm
1.17
Total
6.17
SAMPLE COMPOSITION (>2 mm ct.)
Wood
422
Bark
27
Pitch
10
Rhizome
1
Unknown
7
Total >2 mm
467
WOOD IDENTIFICATIONS Count
Acer saccharum, sugar maple
1
Betula spp., birch
1
Fagus grandifolia, beech
8
Fraxinus spp., ash
6
Pinus spp., pine
P. strobus, white pine
Prunus spp., cherry
Quercus spp., red oak group
4
Total
20
PLANT COMMUNITIES
Northern hardwoods (sugar maple, birch, beech)–white
pine
Bottomland forest (ash)
Oak (northern red oak)
Disturbed woods/thickets (cherry)
Total
Notes: P = present in 0.5-2 mm charcoal.
Features 2 & 5 analyzed 3/20/1991; Feature 2014 analyzed 4/17/2015.
Identifications by Nancy Asch Sidell.

10

3640+/-30

0.98
0.50
1.48

n/a
n/a
0.137

95
13
3
111

10
1
P
11

1
7
3
7
2
20

1
4
5
10

Total
Late Archaic

2
8
11
6
7
1
4
11
50
%
58.00
12.00
22.00
8.00
100.00

 New Investigations of the Cates Farm Site, Vassalboro, Maine

The analysis of wood charcoal from
Cates Farm site (Table 2) revealed a mixture
of wood types in all three features, indicating
that the nearest available deadwood was likely
used for everyday campfires. Classifying
each genus or species according to its usual
habitat (Asch Sidell 2002), the data suggests
that a mixture of plant communities grew at
the outlet of China Lake between 4000-5000
B.P. The assemblage was dominated by
northern hardwoods including northern red
oak. The Cates farm site is outside of the present range of most southern species such as
hickory, black oak, scarlet oak, and bear oak.
The site may be on the edge of the northern
distribution of pitch pine, white oak, and
American chestnut (Maine Forest Service
2008), although none of these wood types
were identified at Cates Farm site.
Given the calcined alewife bone in a
small feature with associated charcoal, radiocarbon dating of the deposit was warranted to
establish the date of harvesting fish at this site
at the outlet of China Lake. Accordingly,
Shaffer submitted a sample of the Feature
2014 cherry wood charcoal to Beta Analytic
for accelerator mass spectrometry radiocarbon
dating. The material was subjected to pretreatment with acid/alkali/acid. The analysis
yielded, after correction for total isotopic fractionation effects, a conventional radiocarbon
age of 3640 ± 30 BP (Beta-413738). Two
sigma calibration using the 2013 INTCAL
program found probability ranges of Cal BC

2130 to 2085 (Cal BP 4080 to 4035) and Cal
BC 2045 to 1930 (Cal BP 3995 to 3880). The
dating indicates that Native Americans were
harvesting alewives probably at Cates Farm
site around 4000 years ago during the Late
Archaic period. This date of 3600 radiocarbon is within the probability range of the earliest Susquehanna tradition occupation of
Maine, after the end of the Moorehead phase.
Discussion and Conclusions
Testing of the footprint of the planned high
tunnel confirmed the presence of significant
archaeological resources of site 38.10. The
single rhyolite flake was found mixed with
historic artifacts in the plowzone. However, a
pit feature (Feature 2014) provided information on site occupation and its biological materials offered significant data on the site’s
economy and dating. With regard to the
NRCS conservation practice, installation of
the high tunnel would entail minimal impacts:
the driving in of anchoring posts. This action
should have no adverse effect on site 38.10,
conditional upon there being no other subplowzone disturbance within the footprint of
the agricultural structure.
Beyond the outcome of the historic preservation review, the recent testing of the Cates
Farm site has added significant information to
knowledge about Late Archaic environment
and subsistence. At 3600 radiocarbon years
ago (about 4000 calendar years), a probable
11

 New Investigations of the Cates Farm Site, Vassalboro, Maine

Susquehanna tradition group of people at the
Cates Farm were harvesting the spring alewife
run, and firing their hearth with wood from
nearby trees that included red oak, cherry, and
white pine. (It is possible that an adult alewife was caught from the lake during the summer, before they returned to the sea. However, it is most likely by far that the bone
marks harvest of the concentrated spring
spawning run into the lake.)
MHPC Feature 2, a probable Vergennes
phase large garbage pit (Trautman and Spiess
1992:40),is radiocarbon dated 5000 ± 70 B.P.
(Beta 44175); and the charcoal (Table 2) includes beech, ash, red oak, birch, and sugar
maple . Feature 2 yielded 171 fragments (2.3
grams) of calcined bone, of which two were
identifiable as muskrat (Ondatra), one as alewife or shad, and eight are unidentifiable
small fish (ibid.: Table 2, p.13). MHPC Feature 5 yielded three calcined bones that could
be identified: large mammal, small fish, and
turtle carapace. Feature 5 charcoal included
pine, birch, beech, red oak, and sugar maple
(Table 2). Thus, during the Late Archaic at
the Cates Farm, people were harvesting fish
runs, other aquatic animals (turtle, muskrat),
and hunting larger mammals, in a hardwooddominated forest.
The Maine forest cover at European settlement is known (at some level of detail)
from surveyors’ records of the trees they encountered as they laid out township lines, socalled “witness trees” (Barton et al. 2012:56-

66). At 1770-1800 A.D. (roughly), the uncut
forest cover in the Waterville–China area fell
within the “Northern Hardwood 2” type (ibid.
62-63). Northern Hardwood forests were
dominated by spruce, beech, and hemlock,
with yellow birch, sugar maple, white pine,
and red oak. However, the townships of the
lower Kennebec River valley immediately
adjacent to the south fell within a “Southern
Oak–Pine” forest cover type dominated by
beech, white pine, hemlock, red oak and white
oak. The boundary between these forest types
was quite abrupt, characterized as an ecotone
(Barton et al. 2012:63). It could be speculated
that warmer climate during the Middle and
Late Archaic may have been characterized by
a more northerly extension of a tree association like the “Southern Oak–Pine” forest.
However, the analysis of wood charcoal from
three features at the Cates Site did not recover
any evidence of “southern” species. The presence of cherry wood in Feature 2014 is an indicator that there may have been openings or
disturbed areas in the forest around 4000 BP.
The ubiquity of red oak wood in all three features suggests that acorns could have been a
food source; however, the absence of charred
acorns may support the idea that the site was
occupied in the spring during the alewife run
rather than in the fall when the acorns would
have been harvested.
While it would take many charcoal identifications from many Native American fire
hearths to “track” the changes in the prehis12

 New Investigations of the Cates Farm Site, Vassalboro, Maine

toric forests of Maine in detail, data such as
these can contribute to understanding ecological change over time. Similarly, animal bone
distributions, especially fish, record change or
continuity in animal species distributions as
the forest cover changed and as water levels
and temperatures rose and fell.

References Cited
Asch Sidell, Nancy
1999 Prehistoric Plant Use in Maine:
Paleoindian to Contact Period. In John
P. Hart, ed., Current Northeast Paleoethnobotany, pp. 191-223. New York
State Museum Bulletin No. 494, Albany.
2002 Paleoethnobotanical Indicators of
Subsistence and Settlement Change in
the Northeast. Chapter 13 in Northeast Subsistence-Settlement Change:
A.D. 700 - A.D. 1300, edited by John
P. Hart, pp. 241-264. New York State
Museum Bulletin 496.

Acknowledgments
We appreciate that Chris Cates has been very
supportive of the archaeological investigations
and other historic preservation studies conducted as part of the NRCS high tunnel project.
MHPC funded Asch Sidell’s
archeobotanical identifications of the charcoal. We thank David Halliwell for comments
improving the manuscript. NRCS Maine’s
State Conservationist Juan Hernandez arranged for funding the AMS radiocarbon dating. Amanda Burton of NRCS Maine assisted
with the fieldwork. Trade names are used
solely to provide information. Mention of a
trade name does not constitute a guarantee of
the product by the U.S. Department of Agriculture nor does it imply endorsement by the
Department or the Natural Resources Conservation Service over comparable products that
are not named. The USDA is an equal opportunity provider and employer.

Barton, Andrew M., with Alan S. White and
Charles V. Cogbill
2012 The Changing Nature of the Maine
Woods. University of New Hampshire
Press, Durham.
Bourque, Bruce J.
1976 Report on an Archaeological Survey of the Lower Kennebec Valley and
Adjacent Areas. Report on file, Maine
Historic Preservation Commission.
Brown, Ann R.
1982 Historic Ceramic Typology with
Principal Dates of Manufacture and
Descriptive Characteristics for Identification. DELDOT Archeology Series
13

 New Investigations of the Cates Farm Site, Vassalboro, Maine

Miller, George L.
2000 Telling Time for Archaeologists.
Northeast Historical Archaeology
29:1-22.

No. 15. Delaware Department of
Transportation.
Faust, Albert P., and Kenneth J. LaFlamme
1978 Soil Survey of Kennebec County,
Maine. U.S. Department of Agriculture, Soil Conservation Service,
Maine.

Miller, Kevin
2015 Battle Revised over Alewives in the
Saint Croix River. Portland Press
Herald, April 11, 2015.

Kelley, Joseph T., Daniel F. Belknap, and
Stefan Claesson
2010 Drowned Coastal Deposits with Associated Archaeological Remains from
a Sea-level “Slowstand”: Northwestern
Gulf of Maine, USA.
Geology:
38(8):695-698.

Petersen, James B.
1991 Archaeological Testing at the Sharrow Site: A Deeply Stratified Early to
Late Holocene Cultural Sequence in
Central Maine. Occasional Publications in Maine Archaeology Number 8.
The Maine Archaeological Society and
the Maine Historic Preservation Commission, Augusta.

Maine Department of Marine Resources
2012 Electronic resource, Maine River
Herring
Fact
Sheet,
http://www.maine.gov/dmr/searunfish/
alewife/, accessed August 10, 2015.

Petersen, James B., Brian S. Robinson, Daniel
F. Belknap, J. Stark and L. K. Kaplan
1994 An Archaic and Woodland Period
Fish Weir Complex in Central Maine.
Archaeology of Eastern North America
22:197-222.

Maine Forest Service
2008 Forest Trees of Maine. Centennial
Edition 1908-2008 (14th edition). Department of Conservation.

Spiess, Arthur E.
1999 Deep Testing on the Kennebec: The
Waterville-Winslow Bridge.
The
Maine Archaeological Society Bulletin
39(2):43-57.
2003 Edwards Dam Removal: 2000 Impoundment Margin Prehistoric Arch-

McGuire, Peter
2015 Alewife Project Could Bring Fish to
China Lake. Maine Sunday Telegram,
March 29, 2015.

14

 New Investigations of the Cates Farm Site, Vassalboro, Maine

aeological Survey.
The Maine Archaeological Society Bulletin 43(1):116.
2004 The Hassan Collection: Otter Creek
Points, and the Archaic on the Kennebec River near Augusta. The Maine
Archaeological Society Bulletin
44(1):1-14.

Spiess, Arthur E. and John P. Mosher
2006 Archaic Period Hunting and Fishing
around the Gulf of Maine. In David
Sanger and M. A. P. Renouf, eds., The
Archaic of the Far Northeast, pp. 383408. The University of Maine Press,
Orono.
Trautman, Liz
1996 A Gouge, a Scraper, an Adze Lying
amongst the Glads: A Further Look at
the Cates Farm Site. The Maine Archaeological Society Bulletin
36(2):25-41.

Spiess, Arthur E. and Leon Cranmer
2000 Edwards Dam Removal: Initial Archaeological Survey Results.
The
Maine Archaeological Society Bulletin
40(2):11-34.
Spiess, Arthur E. and David B. Halliwell
2011 Fish Remains Found at Pre-Columbian Archaeological Sites in Maine:
Freshwater, Diadromous, and Estuarine Fish. The Maine Archaeological
Society Bulletin 51(2):1-26.

Trautman, Elizabeth, and Arthur Spiess
1992 The Cates Farm: Archaic and
Woodland Occupation at China Lake
Outlet. The Maine Archaeological
Society Bulletin 32(1):1-44.
Watts, Douglas
2012 Alewife. Poquanticut Press, Easton,
MA.

Spiess, Arthur E. and Mark Hedden
1999 Late Paleoindian Occupation at the
Waterville-Winslow Bridge. The
Maine Archaeological Society Bulletin
39(2):43-57.
2000 Susquehanna Tradition Activity Areas at the Waterville-Winslow Bridge.
The Maine Archaeological Society
Bulletin 40(1):23-54.

15