Pollen Concentration Analysis of Ancestral Pueblo Dietary Variation

19 Pages·2016·3.93 MB·English

UUnniivveerrssiittyy ooff NNeebbrraasskkaa -- LLiinnccoollnn
DDiiggiittaallCCoommmmoonnss@@UUnniivveerrssiittyy ooff NNeebbrraasskkaa -- LLiinnccoollnn
Anthropology Faculty Publications Anthropology, Department of
2006
PPoolllleenn CCoonncceennttrraattiioonn AAnnaallyyssiiss ooff AAnncceessttrraall PPuueebblloo DDiieettaarryy
VVaarriiaattiioonn
Karl J. Reinhard
University of Nebraska-Lincoln, [email protected]
Sherrian Edwards
Teyona R. Damon
University of Nebraska-Lincoln
Debra K. Meier
PathoEcology Services, [email protected]
Follow this and additional works at: https://digitalcommons.unl.edu/anthropologyfacpub
Part of the Anthropology Commons
Reinhard, Karl J.; Edwards, Sherrian; Damon, Teyona R.; and Meier, Debra K., "Pollen Concentration
Analysis of Ancestral Pueblo Dietary Variation" (2006). Anthropology Faculty Publications. 27.
https://digitalcommons.unl.edu/anthropologyfacpub/27
This Article is brought to you for free and open access by the Anthropology, Department of at
DigitalCommons@University of Nebraska - Lincoln. It has been accepted for inclusion in Anthropology Faculty
Publications by an authorized administrator of DigitalCommons@University of Nebraska - Lincoln.
Published in Palaeogeography, Palaeoclimatology, Palaeoecology 237 (2006) 92–109. Copyright 2006, Elselvier.
doi:10.1016/j.palaeo.2005.11.030. http://www.elsevier.com/locate/palaeo. Used by permission.
Pollen Concentration Analysis of Ancestral Pueblo Dietary
Variation
Karl J. Reinharda,*, Sherrian Edwardsb, Teyona R. Damona, Debra K. Meierc
a School of Natural Resources, University of Nebraska-Lincoln, Lincoln, NE, United States
b Dallas, TX, United States
c PathoEcology Services, Lincoln, NE, United States
* Corresponding author: [email protected] (K.J. Reinhard).
Received March 24, 2003; accepted November 8, 2005.
Abstract
Previous coprolite research on the Colorado Plateau has shown that macrofossils are a useful way of statistically
demonstrating prehistoric dietary variation of Ancestral Pueblos (Anasazi). Up until now, pollen concentration
from human coprolites has not been used for comparative, statistical study. We present here the statistical anal­
ysis of pollen concentration values of coprolites from two Ancestral Pueblo sites, Salmon Ruin and Antelope
House. The data show that although most pollen types do not show statistically significant variation, there are
some types that show how different Ancestral Pueblo populations adapted to plant resources in different environ­
ments. The analysis indicates that future work should focus more on pollen concentration analysis of coprolites.
Keywords: Ancestral Pueblo; Anasazi; Palynology; Pollen concentration; Coprolites; Diet
1. Introduction Ancestral Pueblo dietary practices between two very
different Ancestral Pueblo sites: Salmon Ruin, New
Minnis (1989) demonstrated that coprolite macro­ Mexico and Antelope House, Arizona.
fossils from Ancestral Pueblo (Anasazi) sites were Antelope House in Canyon de Chelly National Mon­
particularly important in identifying culturally­de­ ument, Arizona, and Salmon Ruin near Bloomfield,
fined dietary patterns. “Ancestral Pueblo” refers to New Mexico were excavated with particular attention
the prehistoric precursors of modern Pueblo societ­ paid to recovery of biological remains (Figure 1). Both
ies such as the Hopi, Zuni, and Rio Grande Pueblos. sites were excavated in the “New Archaeology” pe­
Ancestral Pueblo societies were among several pre­ riod in the late sixties and seventies. The focus on the
historic groups that occupied the Colorado Plateau, scientific recovery of biological data was pioneered
a region that includes parts of Arizona, New Mex­ in the excavations of these sites. Coprolites and other
ico, Colorado, and Utah. To date, no researcher has biological remains from both sites have been stud­
attempted to assess the value of pollen concentration ied (Reinhard, 1992, 1996). With regard to other re­
analysis in defining different patterns of Ancestral mains, both are documented by monographs (Irwin­
Pueblo resource use at separate sites. We are taking Williams and Shelley, 1980; Morris, 1986). However,
this opportunity to evaluate the value of coprolite pol­ Antelope House studies are more represented in jour­
len concentration techniques in assessing variation in nal articles and book chapters (Fry and Hall, 1975; Re­
92
Pollen ConCentration analysis of anCestral Pueblo Dietary Variation 93
inhard, 1992, 1993, for review Sutton and Reinhard, paleoethnobotany, it is appropriate that coprolites
1995; Reinhard, 1996). In 1975, volume 41 of the Kiva from these sites continue to be used to develop new
(Journal of the Arizona Archaeological and His­ methods of analysis.
torical Society) was committed to articles concern­ These are particularly good sites to compare. Both
ing Antelope House archaeology. Fewer articles ap­ have Pueblo III Period (AD 1,100–1,300) occupations
peared regarding the biological analysis of Salmon with coprolites. Approximately the same number of
Ruin (Doebley, 1976; Burgess-Terrel, 1979; Doebley, people lived in the Pueblo III occupations of each vil­
1981, 1983; Lentz, 1984). However, the methods used lage (Reinhard, 1996). Both sites have coprolite de­
in the biological analysis were published in mono­ posits that can be sampled to diversify the number of
graph form (Bohrer and Adams, 1977) as well as the individual defecations by separate Ancestral Pueblo
results of the analysis of botanical remains from spe­ people represented by the coprolites (Reinhard, 1996).
cific features (Adams, 1980). A comparative analysis Previous analyses of the coprolites and sites indi­
of available published and unpublished data for the cate that the sites were used year­round and that co­
site was done by Reinhard (1996). Because these sites prolites were deposited year-round (Williams-Dean,
were critical in the development of Ancestral Pueblo 1986; Sutton and Reinhard, 1995). Therefore, there is
94 reinharD, eDwarDs, Damon & meier in PalaeogeograPhy, Palaeoclimatology, Palaeoecology (2006)
no evidence of differential seasonal use of the sites or dicinal use of certain plant taxa. They discovered that
the latrines at the sites. pollen concentration values are particularly compel­
In other ways the sites form a contrast. Antelope ling because they reveal that tremendous amounts of
House is located on the floor of Canyon de Chelly. pollen were consumed by prehistoric Southwestern­
Salmon Ruin is located overlooking the flood plain ers. Pollen concentration values ranged into the mil­
of the San Juan River in open country. The Pueblo III lions of pollen grains per gram of coprolite. Such high
occupation of Antelope House is the final manifesta­ quantities of pollen in human coprolites had been ap­
tion of indigenous occupation of the site since Pueblo preciated previously by only one researcher. Sobolik
I times (AD 700), and since Basket Maker times for (1988) calculated pollen concentration values for copr­
Canyon de Chelly in general, (at least since AD 300). olites from the lower Pecos region of Texas. She also
In contrast, Salmon Ruin was originally built by the found that human coprolites contained large quanti­
Chacoan Ancestral Pueblo as a colony. The Pueblo III ties of pollen. She used pollen concentration values to
occupation of Salmon Ruin is derived from a San Juan interpret the passage of time between the consump­
River Ancestral Puebloans that moved into Salmon tion of pollen and the defecation of pollen. In essence,
Ruin after it was abandoned by the Chacoans in the she interpreted coprolites with very high concentra­
beginning of the Pueblo III Period (AD 1,130). Thus, tion values as evidence of recent consumption of pol­
the Salmon Ruin people were new San Juan occupants len­rich foods and coprolites with low concentrations
and were adapted to the San Juan River environment. as evidence of pollen­rich food consumption many
The Pueblo III people of Antelope House were the de­ days before defecation. Most recently, Reinhard et al.
scendants of a tradition that had lived in Canyon de (2002) used pollen concentration analysis in conjunc­
Chelly for centuries. Importantly, the cultural tradi­ tion with phytolith and macroscopic analysis to re­
tions of the sites were distinct. Antelope House was construct Archaic diet in the northern Sonoran Desert.
a classic Kayenta Ancestral Pueblo village adapted They found that pollen concentration was particularly
to the canyon country of northeastern Arizona. The important in identifying dietary use of yucca, prickly
Pueblo III occupation of Salmon Ruin was of the San pear, mustard family, and grass family. The pollen
Juan Ancestral Pueblo adapted to drier, high mesa concentration data also demonstrate medicinal or di­
country of the San Juan River. Therefore, the sites rep­ etary use of willow and Mormon tea.
resent different traditions associated with different The aim of this study is to evaluate the value of pol­
environments. len concentration in defining dietary differences be­
Pollen analysis has been a central part of coprolite tween two Ancestral Pueblo sites.
research from the earliest studies in North America
(Martin and Sharrock, 1964). For most of the history of 2. Materials and methods
coprolite pollen research, pollen data have been pre­
sented as percentage expression of pollen taxa pres­ The coprolite sampling strategies for the sites and
ent in studied coprolites. In the last decades, a newer determination of human origin were detailed by Re­
method of presenting pollen data was applied to co­ inhard (1996). The goal of the strategy was to diver­
prolite pollen analysis. This is the pollen concentra tion sify the samples so that many defecations by separate
method that allows one to calculate the approximate humans were sampled. At Antelope House, this was
number of pollen grains per unit measure of copro­ accomplished by taking single coprolites from several
lites. This method was reviewed by Maher (1981). Ma­ separate, discrete, and dated latrines. At Salmon Ruin,
her presents methods of calculating the numbers of only one latrine was sampled. It was, however, a very
pollen grains per gram of sediment using the follow­ large, stratified deposit of coprolites of which an es­
ing formula: timated 10,000 were excavated and curated. One co­
prolite was taken from alternate 10 cm levels in alter­
Pollen concentration = ((p/m) × e) / w nate 1 m grids. The sampling was done by Reinhard
and Meier.
p pollen grains counted Ultimately, 180 coprolites from Antelope House
m marker grains counted (Reinhard, 1992) and 112 coprolites from Salmon Ruin
e number of exotic marker pollen grains added were selected for analysis. The macrofloral remains
w weight or volume of sediment from all of these were analyzed (Sutton and Reinhard,
1995; Reinhard, 1996). All of these were analyzed mi­
Researchers began to apply this method to human croscopically for parasites before pollen processing
coprolites. Reinhard and colleagues (1991) used pollen (Reinhard, 1992). In the parasite analysis, differen­
concentration data to develop interpretations of me­ tial diagnosis of Equisetum spores versus Populus pol­
Pollen ConCentration analysis of anCestral Pueblo Dietary Variation 95
len was done based on the identification of elators on For statistical analysis, SAS was used for calculation
spores. A subsample was analyzed for phytolith con­ of descriptive statistics, chi square values, and Wil­
tent (Reinhard and Danielson, 2005). Finally, Edwards coxon analysis. The NPAR1WAY procedure was used
and Reinhard analyzed the pollen from 52 coprolites, to determine Wilcoxon scores (rank sums) for pollen
26 from each site. counts by variable site.
Reinhard (1993) published his comparative obser­
vations of coprolite pollen recovery from very lim­ 3. Results and analysis
ited chemical processing to extensive chemical pro­
cessing. With regard to Ancestral Pueblo coprolites Sixteen pollen categories were chosen for statisti­
from Utah, New Mexico, and Arizona, Reinhard cal analysis. Whole maize, broken maize, and total
found that equal results were achieved with both ex­ maize categories were chosen to determine if there
tremes. was a difference between the sites in the consumption
One gram fragments of Antelope House coprolites of ground grain, assuming that broken maize grains
were selected for analysis. One gram fragments from resulted from grinding (Bryant and Morris, 1986).
most Salmon Ruin coprolites were available for study. During macroscopic analysis, the terminal nodes im­
However, some fragments were only 0.75 or 0.5 g. To mediately proximal to the strobili of Equisetum (horse­
each sample, one Lycopodium spore tablet containing tail) were found in several Antelope House coprolites
11,400 spores was added. All fragments were taken (Sutton and Reinhard, 1995). Therefore, we thought it
from the interior of the coprolite. The coprolites des- would be of interest to compare the frequency of Eq-
cribed in this paper were processed through exten­ uisetum spores between the sites. Similarly, one An­
sive chemical treatments following Williams-Dean telope House coprolite was composed of fiber with
(1986) with one exception. The samples were rehy­ thousands of Typha latifolia pollen grains (Sutton and
drated in 0.5% trisodium phosphate for 48 h. After re­ Reinhard, 1995). This coprolite was so rich in pollen,
hydration, macroscopic remains were screened from that it actually appeared yellow. Therefore, Typha was
the microscopic remains and the microscopic resi­ a logical choice for comparative pollen analysis. The
dues were washed three times in distilled water. The macroscopic analysis showed a difference in Rhus (su­
sediments were treated in approximately 40% hydro­ mac) and Phaseolus vulgaris (bean) consumption with
chloric acid. After three distilled water washes, the these plants more commonly occurring in Salmon
sediments were left for 24 h in approximately 70% hy­ Ruin coprolites. Phytolith analysis showed that Opun-
drofluoric acid. The samples were then washed re­ tia was more frequently eaten at Antelope House (Re­
peatedly in distilled water until the supernatant was inhard and Danielson, 2005). Therefore, we chose
clear. After water washes and one glacial acetic acid Rhus, Phaseolus, and Opuntia as comparative catego­
wash, the residues were treated with a 20 min acetol­ ries. Cleome pollen was common in the coprolites and
ysis treatment at 100°C. After one glacial acetic acid the high frequencies of this type begged exploration.
wash, the samples were then washed repeatedly in In previous analyses of coprolites, we found high
distilled water until the supernatant was clear. Finally spine Asteraceae to be common and therefore, we
and unlike Williams-Dean (1986), the sediments were chose this category for analysis. Finally we chose to
treated in 0.5% KOH for 2 min and washed in distilled evaluate a number of anemophilous types to gain an
water three times. This was done to facilitate staining idea of how much pollen could be ingested from the
in basic fuchsin. The samples were then transferred ambient environment and if it was possible to sort out
to 1 dram vials and stored in glycerine. At least 200 dietary use of these types from ambient contamina­
pollen grains were counted for each sample, and up tion. The anemophilous types chosen for study were
to 1,000 grains were counted for some samples. Pol­ low spine Asteraceae, Cheno–Am (Chenopodiaceae/
len types were identified with reference collections of Amaranthaceae), Juniperus (juniper), Pinus (pine), and
Colorado Plateau pollen samples. Single pollen grains Poaceae (grass family).
and pollen aggregates were counted and tabulated. The pollen concentration values for Salmon Ruin
We noticed that many maize pollen grains were bro­ and Antelope House are presented in Tables 1 and 2
ken, shredded, or fragmented. These grains were con­ respectively. The comparative descriptive statistics for
sistent with those described by Bryant and Morris sixteen select taxa are presented in Table 3. The first
(1986) associated with grinding stones. We counted stage of the analysis was comparison of the frequency
broken maize grains separately in order to determine of occurrence of the categories between the sites. Chi
whether there was significance in this observation. square analysis (Table 4) showed that the frequency
Only maize annuli were counted for the fragmented differences were significant at the 0.05 level only for
maize grains. Typha and Equisetum. Both of these taxa were more
96 reinharD, eDwarDs, Damon & meier in PalaeogeograPhy, Palaeoclimatology, Palaeoecology (2006)
Pollen ConCentration analysis of anCestral Pueblo Dietary Variation 97
98 reinharD, eDwarDs, Damon & meier in PalaeogeograPhy, Palaeoclimatology, Palaeoecology (2006)
Pollen ConCentration analysis of anCestral Pueblo Dietary Variation 99
100 reinharD, eDwarDs, Damon & meier in PalaeogeograPhy, Palaeoclimatology, Palaeoecology (2006)