The Effects of Food Processing on the Archaeological Visibility of Maize: An Experimental Study of Carbonization of Lime-treated Maize Kernels 12  Research Communica on  employing a charring technique that allows for complete carbonization of the kernel, without damage. Hence, this research adds to the body of data on ancient maize processing techniques by specifically exploring how alkali processing affects the archaeological preservation of different maize varieties (Martinez-Bustos et al. 2001). The process of alkali cooking, known as hominy production in the Eastern Woodlands and nixtamalization in Mesoam- erica, was widely used by societies living throughout Mesoamerica and North America. I first address the origins and use of alkali processing in the Americas. This is followed by a brief consideration of quantifica- tion methods and lab procedures, which is followed by results of the experiments and discussion of their implications for maize-variety identification, pro- cessing method identification, and general importance in New World archaeology. Alkali Processing of Maize The purpose of cooking maize with an alkali sub- stance is to lengthen its storage life, to increase its nutritional content by changing the chemical and physical composition of the kernels, and to facilitate the removal of the pericarp (Martínez-Bustos et al. 2001). Unprocessed, maize is deficient in niacin and Introduction The effects of ancient processing on the archaeologi- cal visibility and recovery of maize is important for reconstructing past subsistence practices in the New World. Because maize was a prominent cultivar in the Americas, it is important to determine the varieties people grew and the processing methods employed. Previous studies by Goette et al. (1994), King (1987), and Pearsall (1980) addressed this issue through recreating ancient processing techniques (e.g. boiling, sprouting, and parching) to see how these methods affected preservation in the archaeological record. However, these studies were broad in scope and used charring techniques that provided inconclu- sive results. Because of severe distortion to kernel morphology, measurements of kernel shape and size were inaccurate and did not provide reliable results regarding how processing affects carbonization and preservation. As a result, past studies were unable to make direct comparisons between modern carbonized kernels and archaeological carbonized kernels to determine the processing techniques used. This study compares archaeological and modern samples to determine how alkali processing alters phenotypes of New World maize varieties, as well as to establish parameters for identifying alkali processed kernels The Effects of Food Processing on the Archaeological Visibility of Maize:  An Experimental Study of Carboniza on of Lime‐treated Maize Kernels  Caroline Dezendorf  Author address: Interna onal Studies Program, University of Oregon, 175 Prince Lucien Campbell Hall, Eugene, OR 97403.  dezendor@uoregon.edu  Received: September 26, 2012  Volume: 4:12‐20  Published: January 19, 2013  © 2013 Society of Ethnobiology  Abstract: This paper explores the effects of maize processing on the carboniza on and preserva on of maize kernels in the  archaeological record. The shi  to processing maize with lime (known as hominy produc on in the Eastern Woodlands and  nixtamaliza on  in Mesoamerica)  in ancient  mes had the effect of making maize more nutri ous through  increasing the  availability of calcium, niacin, dietary fiber, and essen al amino acids.  Less understood is how this process of cooking maize  in  a  lime  solu on  affects  the  archaeological  preserva on  of  maize;  if  there  is  a  clear  difference  in  the  archaeological  signature of maize remains that are and are not processed this way, then this process may be iden fiable in the archaeologi- cal record. To this end, an experiment was constructed analyzing the varia on in size between dried and alkali processed  maize kernels before and a er carboniza on. Results indicate that alkali processed maize kernels are less likely to fragment  during carboniza on. Key Words: maize processing, nixtamaliza on, Mesoamerica, paleoethnobotany, carboniza on   13  Research Communica on  the essential amino acids lysine and tryptophan; however, alkali processing enhances the potency of these nutrients (King 1987). The traditional process involves the use of alkali cooking, steeping, and washing to remove maize skin from the kernels before consumption or storage (Martinez-Bustos et al. 2001). Martinez-Bustos et al. (2001) state that cooking with an alkali substance (e.g., wood ash) allows for increased water uptake, thus resulting in kernel expansion. The amount of water uptake depends heavily on the maize genotype and processing conditions, such as water temperature and time soaked and cooked (Martinez-Bustos et al. 2001). However, depending on how the maize variety reacts to the process, alkali-processed maize kernels could be differentially preserved through carbonization compared to unprocessed carbonized kernels. Four varieties of maize representing modern descendants of archaeological varieties—also known as heirloom varieties―were selected for experimenta- tion to determine how different maize varieties are affected by alkali processing. It is impossible to obtain modern maize with the exact genetic make-up of their archaeological ancestors, due to evolution; thus, the kernels were acquired from Seeds of Change (www.seedsofchange.com) based on geographic origin and cultural connections (i.e. the historical signifi- cance of the heirloom variety in a given region). The four types selected were Oaxacan Green, Anasazi Flour, Hopi Pink, and Hickory King. Both the Oaxacan Green (originating from the Zapotec Indians in Southern Mexico and green in color) and the Hickory King (originating in the southeast United States and yellow in color) are dent varieties. Dent maize is characterized by a starchy endosperm in the middle, extending towards the top of the kernel that is surrounded by a corneous endosperm along the sides (Sturtevant 1898). Hickory King is known to be used for hominy in the southeastern United States (www.victoryseeds.com). In contrast, Anasazi Flour (a multi-colored maize) and Hopi Pink are soft-flour varieties from the southwestern United States. The flour varieties are considered soft due to the absence of corneous endosperms and the lack of indentation (Sturtevant 1898). It is difficult to make a direct comparison between archaeological specimens and contemporary races of maize even when the evolutionary relation- ships are well understood because maize is constantly evolving due to human selection (Benz 1994). Therefore, there is little information available that allows archaeologists to make comparisons of maize kernels over large areas and through time. Neverthe- less, by studying farmer or heirloom varieties, for which change is slower compared to many other varieties, archaeologists can create a statistical method of analysis to demonstrate correlations of size between modern and ancient varieties (Blake and Cutler 2001). As most archaeological maize is carbon- ized, if the effects of alkali processing can be deter- mined, it may be possible to better characterize archaeological varieties. According to King (1987), there are five kernel measurements that can be used to determine variety; these include angle, length, width, thickness, and distance from base to widest part of the kernel. Several studies (e.g., Pearsall 1980; Goette et al. 1994; Blake and Cutler 2001) have attempted to determine variety using kernel angle to estimate the number of rows on each cob. However, this measure- ment does not take into account distortion and shrinkage due to charring or processing. Blake and Cutler (2001) suggest that up to 40% distortion in characteristics can occur during the carbonization process, which alters the kernel angle. Thus, kernel angle is not an accurate tool for determining varieties and the most accurate results are based on determin- ing pre- and post-carbonization length, width, and thickness (King 1987). Given these challenges, the purpose of this experimental study is to: (1) determine the best way to identify alkali processing of maize in the archaeobotanical record and (2) understand how the alkali process affects different maize varieties. Previous Research As suggested by King (1987), processing method plays a role in determining charred kernel shape. Under- standing the differences in carbonized kernel mor- phology is essential for characterizing different varieties. In terms of alkali processing, I draw on several previous studies. The first study, conducted by Goette et al. (1994), examined three different races of Andean maize using techniques of toasting, sprouting, and boiling. The second study, based in North America and conducted by Francis King (1987), describes how alkali processing alters the kernel shape. Both experiments conclude that most maize recovered from archaeological sites was likely boiled with wood ash (Goette et al. 1994; King 1987). Charred kernels with endosperm extrusion (when the endosperm expands greatly, causing a fragile and extreme distortion to the kernel) are unlikely to 14  Research Communica on  survive in the archaeobotanical record. Therefore, maize kernels processed by sprouting or toasting, which have high percentages of extrusion, are unlikely to survive, especially compared to boiled kernels, which only have an extrusion percentage of 10-15% (Goette et al. 1994). Goette et al.’s (1994) findings support ethnographic data suggesting the process of boiling maize with wood ash was a widespread practice throughout North and South America. Because Goette et al. (1994) focused on only three varieties indigenous to Peru, it is important to expand experimentation to other varieties so as to provide an accurate representation of processing and preservation throughout the New World (Goette et al. 1994). Hence, further knowledge of how alkali processed kernels react to carbonization will help archaeologists determine the processing method used and the number of different varieties at a given site based on morphological characterization (Goette et al. 1994). Because most recovered archaeobotanical remains are charred, it is difficult to distinguish maize varieties and processing techniques based on morphology alone. Past experiments used charring methods in an attempt to recreate archaeological maize and to understand how carbonization affected morphology. Often, these studies provided inconclusive results due to the nature of particular charring methods that left kernels indistinguishable. For example, Cutler and Blake (1973) suggest that kernels carbonized loose (i.e. not on the cob) become too distorted to provide any type of meaningful measurements. However, other scholars indicate that kernels can be charred not on the cob, when packaged tightly together, which avoids extreme distortion to the kernel shape (Goette et al. 1994; King 1987). Nevertheless, because previous charring experiments were conducted under condi- tions of too rapid or too high heat, the morphology of the charred kernels were too distorted and provided inaccurate and unreliable measurements for determin- ing kernel size and shape (King 1987; Pearsall 1990). In past experiments parching techniques were used to obtain measurable results (Goette et al. 1994; King 1987; Pearsall 1990). However, parching likely does not provide analogical realism as it does not cause the same changes in morphology or the representation of archaeological kernels that is caused by charring. For the current experiment it was important to develop and use a method of charring that would preserve the features (length, width, and thickness) and shape of the kernels so as to make them comparable to archaeological specimens. Materials and Methods To determine how alkali processing and carbonization affects the morphology of kernels, it was necessary to perform a laboratory experiment to determine how these processes affect kernels of the four selected varieties of maize. Based on experimental design from Goette et al. (1994) and King (1987), an experiment was constructed to analyze size variability among maize kernels that were a) dried, b) alkali processed, c) dried and carbonized, and d) alkali processed and carbonized. Sample size varied from 40-45 kernels per variety. Initially, each dried kernel was photographed and measured using a stereo-microscope with a camera attachment. I recorded kernel weight, length, width, and thickness, as these are morphological characteris- tics that relate to reproduction and that are least affected by environmental variability in phenotype (Goodman and Paterniani 1969). The next step involved a 50% random selection of each variety to be processed in an alkali solution. In order to use historically appropriate methods and follow conven- tion of previous alkali processing experiments, ten pounds of uncontaminated oak, maple, and ash wood were burned to produce ash for the alkali solution. Three cups of the hardwood ash were boiled in six quarts of water for one hour to achieve a pH of 10 (Goette et al. 1994). The water was then sieved and divided between four pots, one for each maize variety, and the maize was soaked for fourteen hours (Martinez-Bustos et al. 2001). After soaking, the maize was cooked at 85º Celsius for one hour, at which point the pericarps began to loosen. The kernels were then rinsed under running water and rubbed together to remove seed coats and points of attachment. After drying, the alkali-processed kernels were re-measured. The next step was carbonization. Because research by Blake and Cutler (1973) suggests that kernels carbonized loose will cause distortion, kernels were placed in tin-foil packets before carbonization. In total there were eight tin-foil packets that were carbonized—two for each variety with one containing the unprocessed kernels and the other containing the alkali-processed kernels. The kernels were cooked in a muffle furnace at 180-190ºC, conditions that represent a low-burning fire, for one hour and then were removed from the furnace to cool for three minutes (Werts and Jahren 2007). They were then cooked for one more hour at which point smoking ceased, which is indicative of complete charring. The purpose of the 15  Research Communica on  double cooking method was to mitigate issues that previous studies experienced—these include carboniz- ing kernels in too high of heat or too rapidly. Because this research is experimental, temperature and cooking time were controlled. However, archaeological kernels would not have been exposed to consistent heat, as fire temperatures and surrounding soil temperatures vary (Werts and Jahren 2007). During the initial phase of the carbonization process, the unprocessed kernels cracked, at an average rate of one to two pops per five minute period, exhibiting splitting and swelling. After one hour, these unprocessed kernels became sur- rounded by a foamy black matrix. When carboniza- tion was complete, all kernels were re-measured using the same methods as prior to processing. Dependent t -tests are used to assess differences in shrinkage among dried, alkali-processed, and carbonized kernels. Results By using a controlled charring environment, I was able to carbonize specimens that exhibit similar morphologies to archaeological specimens (including those found at sites in the Midwestern United States1), which allows insights into how processing affects maize kernel taphonomy. The metric dimensions of the carbonized alkali-processed kernels were greater than those of non-alkali-processed kernels. There are substantial differences between the (a) dried, (b) alkali processed, (c) unprocessed carbonized, and (d) alkali processed carbonized kernels. There were clear differences in thickness and width amongst the varieties. The increase in thickness between dried (a) and alkali-processed-carbonized (d) ranged from 21.64% for Anasazi Flour to 71.38% for Hickory King. In addition, the lengths of the kernels decreased in size after alkali processing and again after carboni- zation; this decrease ranged from 3.09% for Hickory King to 22.05% for Hopi Pink. Goette et al. (1994) report similar results, indicating that the decrease in moisture causes vertical contraction of kernels, whereas increasing internal pressure causes horizontal expansion. These differences are best demonstrated visually with scatterplots. Note that the majority of alkali- processed-carbonized kernels show an increase in width and thickness (Figure 1), when compared to the (non-carbonized) dried and alkali-processed kernels. In addition, the unprocessed-carbonized kernels show an increase in thickness, but not width, which indicates that carbonization is a contributing factor to increased thickness. It is thus logical to conclude that the increase in internal pressure that occurs during carbonization forces the kernels to expand in thick- ness; however, because of decreased moisture content in the absence of alkali processing, other factors, such as length and width, appear not to increase under carbonization alone. Therefore, the magnitude of width expansion is likely dependent upon cooking method. This interpretation is supported by visual inspection of Figure 2, which plots length by width for all four varieties under different experimental conditions. By examining the scatterplots, it is evident that the alkali-processed and the alkali-processed- carbonized kernels exhibit the greatest shift in width when compared to the dried maize and unprocessed- carbonized kernels. The majority of the alkali- processed Hopi Pink kernels and Anasazi Flour Kernels show the largest increase in width after processing, and exhibit a slight decrease in width with carbonization. The fact that the alkali-processed kernels exhibit greater width than the alkali-processed -carbonized kernels indicates a decrease in kernel moisture content caused by carbonization. Unless kernels are alkali processed, they should not have significant changes in width. The use of dependent t-tests allows for statistical analysis of shrinkage in kernel dimensions. A compar- ison of the dried kernels (a) with the alkali processed kernels (b) was beneficial to determining the impact that alkali processing has on kernel shape. Tests on width measurements between these pairs (for all varieties) yield p-values less than 0.05, indicating significant differences in width—in other words, alkali processing leads to significant kernel swelling along this dimension (Table 1). There is also a significant difference in width between unprocessed carbonized kernels (c) to alkali processed carbonized kernels (d) (p-value < 0.001). Dependent t-tests were also used to determine statistical differences in thickness (Table 1). For all varieties, it was determined that thickness increases with carbonization. Additionally, unprocessed carbonized kernels (c) and alkali processed carbonized kernels (d) were compared in order to determine if the measurement of thickness could be used to determine the different processing methods in archaeological kernels. Dependent t-tests demonstrate that the differences are significant; thus, for archaeo- logical specimens, comparisons of the width and the thickness of kernels, assumed to be the same variety, 16  Research Communica on  Figure 1. Results from sca er‐plots of width vs. thickness for Anasazi Flour, Hickory King, Hopi Pink, and Oaxacan Green  maize kernels.   17  Research Communica on  Figure 2. Results from sca er‐plot of length vs. width Anasazi Flour, Hickory King, Hopi Pink ,and Oaxacan Green maize ker‐ nels.  18  Research Communica on  can allow researchers to determine whether or not alkali processing occurred before carbonization. Kernels that exhibit increased thickness and width indicate alkali processing and carbonization whereas, kernels that only exhibit increased thickness suggest carbonization alone. Additionally, because data suggest that the widths and thicknesses for each variety fall within certain ranges, researchers could separate kernels into varieties. Specifically, dent varieties of maize display larger width and thickness dimensions than flour varieties. After carbonization, the median width for the alkali processed Anasazi flour kernels was 8826.5 µm and the median width for the alkali processed Hopi Pink was 9713.0 µm. In contrast the processed and carbonized Oaxacan Green dent measured 10994.9 µm and the Hickory King dent measured 1499.36 µm. Results for thick- ness demonstrate a similar pattern with the median Anasazi Flour and Hopi Pink, measuring 5829.1 µm and 7079.1µm, respectively, while median thickness for Oaxacan Green and Hickory King were 6607.1 µm and 7710.5 µm, respectively. Both measurements indicate that flour varieties will be smaller in size after carbonization than dent varieties. After experimental carbonization, there was a large phenotypic difference between the kernels that were boiled in the alkali solution and then carbonized compared to those that were only carbonized. The reason for this difference is that the effect of heating on kernels is directly correlated to endosperm composition (King 1994). Because the four varieties were composed mainly of floury endosperms, they were more likely to swell and split due to intense pressure build-up in the early stages of carbonization (King 1994). The unprocessed kernels expanded and split, making them unrecognizable, while the alkali- processed kernels maintained their shape and struc- ture. The majority of Hickory King, Oaxacan Green, Hopi Pink, and Anasazi Flour unprocessed kernels split and became globulated, which indicates that the starches oozed out the pericarp due to internal pressure that produced foamy black matrices. The kernels also became very brittle—a clear indicator that unprocessed kernels are not good candidates for archaeological preservation as they would likely suffer significant mechanical damage, leading to higher rates of fragmentation. In contrast, the alkali processed kernels appeared very similar to archaeological specimens (e.g., from Myer-Dickson and Roskamp sites located in the Central Illinois River Valley in the midwestern United States). A few of the Oaxacan Green kernels cracked along the exterior, which is likely due to not being soaked long enough in the alkali solution. Neverthe- less, all of the kernels were identifiable and durable. It is important to note that unlike some archaeological kernels, the embryos of the kernels used in the experiment stayed attached after carbonization, which Note: A = dried ; B =  alkali processed ; C = unprocessed carbonized ; D =  alkali processed carbonized ; * indicates sta s cal significance  Maize Type  Pairs for Ker‐ nel Width  t  df  p‐ value     Pairs for Kernel  Thickness  t  df  p‐value  Anasazi Flour  A / B  3.855  50.5  0.000*  Anasazi  Flour  A / C  8.301  20.4  0.000*     C / D  0.169  28.3  0.867     C / D  4.723  26.3  0.000*  Hickory King  A / B  3.541  51  0.001*  Hickory  King  A / C  13.661  19.5  0.000*     C / D  4.813  26  0.000*     C / D  4.881  27.8  0.000*  Hopi Pink  A / B  5.377  38.1  0.000*  Hopi Pink  A / C  7.837  24.1  0.000*     C / D  3.817  31.7  0.001*     C / D  2.195  29.5  0.036*  Oaxacan Green  A / B  5.336  40.4  0.000*  Oaxacan  Green  A / C  14.24  14.5  0.000*     C / D  4.926  23.2  0.000*     C / D  8  16.9  0.000*  Table 1. Results of two‐sample t‐tests with separate variances for Anasazi Flour, Hickory King, Hopi Pink, and Oaxacan  Green maize kernels.  19  Research Communica on  is most likely due to the effects of uniform tempera- ture in the controlled carbonization environment. The lack of distortion in the alkali-processed kernels likely relates to the fact that boiling softens the endosperm, thus allowing the kernel to swell without splitting. Discussion and Conclusion It is evident from these results that the distinguishing characteristics of carbonized maize are not only based on the genotypic variety but also on the processing mechanisms which the kernels undergo (Goette et al. 1994). Upon carbonization, most of the alkali processed kernels were similar in appearance to kernels recovered at archaeological sites in the midwestern U.S. (Myer-Dickson, Roskamp, Lamb)i, which do not have embryos and are broad and crescent shaped (King 1994). Therefore, it can be assumed that various Native American groups were using a method of alkali processing, which results in better archaeological preservation of the kernels. However, the use of alkali processing creates a preservation bias at archaeological sites due to the fact that it is primarily alkali processed kernels that remain complete, compared to unprocessed kernels which are more brittle and tend to fragment (King 1994). Because the studies indicate that unprocessed kernels become distorted when carbonized, the probability of finding identifiable unprocessed complete kernels is slim compared to finding kernels that are alkali processed (Goette et al. 1994; King 1994). Additional- ly, certain varieties, based on endosperm composition, were more likely to be alkali processed than others, and thus, the varieties of recognizable maize found at archaeological sites favor those that underwent alkali processing, which would most likely be dent and flour varieties. Due to the fact that there have not been many studies conducted on alkali processing, further research is needed to determine archaeological varieties that underwent processing. What this study does conclude is that alkali processed kernels, which are not brittle and are less likely to fragment, are more readily found at archaeological sites than unprocessed kernels. Through the use of statistical analysis that examines kernel morphology, researchers will be able to categorize and determine the number of varieties at a site based on measurements of length, width, and thickness. It is also important to note that there are prob- lems based on developing methods of measurement and of statistical analysis that can be applied to maize varieties from different sample groups. Because of genetic differences, it is difficult to make direct comparisons between the samples created in a laboratory and those formed in the archaeological record (Goette et al. 1994). However, understanding of genealogy and varietal relationships can help mitigate this problem. In addition, because every variety of maize reacts differently to alkali processing and, therefore, the results for one variety will be different than those of another variety, different varieties can be grouped based on statistical differ- ences. Finally, this study focused on flour and dent varieties; we still do not understand how flint, sweet, or popcorn varieties (which have different endosperm compositions) are altered by alkali processing. Nevertheless, this study does provide conclusive evidence that alkali processed kernels, when carbon- ized, will demonstrate increased widths, increased thickness, and loss of pericarps and points of attach- ment (indicative of alkali processing). If kernels were not alkali processed, they would not lose their pericarps, and therefore, would have different phenotypic compositions than alkali processed kernels. Future studies must be conducted in order to determine which archaeological maize samples were alkali processed. The current study demonstrates that there are morphological similarities between modern alkali processed, carbonized kernels and archaeologi- cal kernels; however, studies analyzing the chemical composition of archaeological kernels are still needed. It would also be beneficial to conduct burial studies to analyze how modern alkali processed, carbonized varieties, placed underground for an extended period of time, react to environmental pressures. The use of burial studies would allow researchers to determine how local environmental conditions affect composi- tion of the alkali processed kernels to determine if more phenotypic or chemical changes occur. As suggested by modern ethnographic data, there is a high correlation between societies that cultivate and consume maize and those that use alkali pro- cessing (King 1987). Additionally, we know that alkali processing diffused from Mesoamerica to North America, but further research is necessary to under- stand when diffusion occurred (King 1987). Future studies must also determine when different groups in Mesoamerica and North America first used alkali processing. To do this, analysis of a broader range of maize varieties, including archaeological kernels from different time periods, needs to occur. Nevertheless, 20  Research Communica on  this study provides clear evidence of the importance of alkali processing for the preservation of archaeo- logical maize and for the identification of different maize phenotypes. Understanding how alkali processing affects kernel morphology will allow archaeologists to formulate a statistical method for determining processing techniques and maize varieties in the archaeobotanical record. Acknowledgements This research was originally written as my honor’s thesis at the University of California, Santa Barbara (UCSB) in 2011 and was presented at the 2011 Society of Ethnobiology Conference under the direction of Dr. Amber VanDerwarker. Archaeo- logical and experimental aspects of this research were conducted at UCSB. I would like to thank Dr. VanDerwarker, Dr. Greg Wilson, and Dana Bardolph for editorial comments and for support and encouragement of my research. Two anony- mous reviewers provided helpful comments and suggestions. Declarations Permissions: Not applicable. Sources of funding: None declared. Conflicts of interest: None declared. References Cited Benz, B.F. 1994. Can Prehistoric Racial Diversifica- tion Be Deciphered from Burned Corn Cobs? In Corn and Culture in the Prehistoric New World, edited by S. Johannessen and C.A. Hastorf, pp. 23-33. Westview Press, Boulder, Colorado. Bird, R.M. and M.M. Goodman. 1977. The Races of Maize V: Grouping Maize Races on the Basis of Ear Morphology. Economic Botany 31:471-481. Blake, L.W. and H.C. Cutler. 2001. Plants from the Past. The University of Alabama Press, Tusca- loosa. Cutler, H.C. and L.W. Blake. 1973. Plants from Archaeological Sites East of the Rockies. Missouri Botanical Garden, St. Louis. Goette, S., M. Williams, S. Johanneseen, and C.A. Hastorf. 1994. Towards Reconstructing Ancient Maize: Experiments in Processing and Charring. Journal of Ethnobiology 14:1-21. Goodman, M.M. and E. Paterniani. 1969. The Races of Maize: III. Choices of Appropriate Characters for Racial Classification. Economic Botany 23:265 -273. King, F. 1987. Prehistoric Maize in Eastern North America: An Evolutionary Evaluation. Unpublished Doctoral Dissertation, Department of Agronomy, University of Illinois, Urbana. King, F. 1994.Variability in Cob and Kernel Charac- teristics of North American Maize Cultivars. In Corn and Culture in the Prehistoric New World, edited by S. Johannessen and C.A. Hastorf, pp. 35-54. Westview Press, Boulder, Colorado. Martinez-Bustos, F., H.E. Martinez-Flores, E. Sanmartin-Martinez, F. Sanchez-Sinencio, Y.K. Chang, D. Barrera-Arellano and E. Rios. 2001. Effect of the Components of Maize on the Quality of Masa and Tortillas During the Traditional Nixtamalisation Process. Journal of the Science of Food and Agriculture 81:1455-1462. Pearsall, D.M. 1980. Analysis of an Archaeological Maize Kernel Cache from Manabi Province, Ecuador. Economic Botany 34:344-351. Smith, B.D. 1995. The Emergence of Agriculture. Scien- tific American Library, New York. Sturtevant, E.L. 1898. Varieties of Corn. USDA Experiment Station, Bulletin 137. Werts, S.P. and A.H. Jahren. 2007. Estimation of Temperatures Beneath Archaeological Campfires Using Carbon Stable Isotope Compostion of Soil Organic Matter. Journal of Archaeological Science 34: 850-857. Biosketch Caroline  Dezendorf  is  currently  pursuing  her  Master’s  Degree  in  Interna onal  Studies  at  the  University  of  Oregon. Her current research focuses on issues of food  jus ce and La no immigra on.   Notes 1 These samples are currently being analyzed in Dr. Amber VanDerwarker’s lab at UC-Santa Barbara.