21 Dental Anthropology 2022 │ Volume 35│ Issue 01 Tooth enamel is the densest, most resilient tissue in the human body (Hillson, 1996). As a result, hu- man teeth typically can survive a wide range of environments, making them a rich source of infor- mation for bioarchaeologists gathering data on hu- man behavior. Indeed, the importance of the denti- tion in bioarchaeology relates to the fact that it in- forms on human evolution, diet, growth and devel- opment, migration, identity, and disease (Scott and Turner, 1988; Hillson, 1996). Since the oral cavity has direct contact with both the external and inter- nal environment, examination of oral disease in the dentition enhances our understanding of the differ- ences in foodways between and within cultures. Dental disease provides significant information concerning ancient diet and cultural practices, as well as the influence of diet on pathological condi- tions of the dentition (Konig, 2000; Moynihan, 2005). Dental disease is sometimes used as a proxy for understanding oral health, but the inconsisten- cy in defining the term health has led researchers to move away from the umbrella term that includes unknowable factors (e.g., psychosocial aspects) and instead focus on dental disease as indicated by spe- cific conditions (e.g., dental caries; Pilloud and Fancher, 2019). Although teeth are valuable indicators of dis- ease and life history, as well as a source of demo- graphic and cultural data, several studies highlight the prevalence of sample bias arising from ante- mortem and postmortem teeth loss (e.g., Lukacs, 1995; Erdal and Duyar, 1999). Loose or missing teeth are extremely common in bioarchaeological samples, and a review of the literature has shown inconsistent methods in dealing with the conse- quent bias during data collection and analysis. The study of dental pathology is further complicated by the varying preservation rates of the multiple Gaps in Information: What Missing Teeth Mean in Bioarchaeology Laura E. Cirillo 1* and Eric J Bartelink 2 1 Department of Anthropology, University of Nevada, Reno 2 Department of Anthropology, California State University, Chico ABSTRACT Previous bioarchaeological analysis of postmortem tooth loss (PMTL) has failed to recog- nize the potential influence of diseased dental tissue on tooth retention after death. Because tooth loss from a traditional taphonomy prospective is treated simply as missing data, demographic studies are potentially influenced by underestimations of disease prevalence. To investigate the association of tooth loss and dental disease, data on the pathological conditions observed in the tissues were collected on a sample of teeth from 771 individuals. By analyzing the evidence of disease in the bone and dental tis- sues immediately surrounding empty alveolar sockets suggestive of PMTL, trends in the presence of diseased tissue and retention of a tooth emerged. When compared to teeth retained after death, PMTL sockets were 15.3% less likely to retain neighboring teeth and 21.5% less likely to have neighboring teeth that showed no signs of carious or periapical lesions. The results suggest that the traditional explanation of susceptibility to loss due to the exposure and morphology of single-rooted, anterior teeth does not sufficiently explain the causes of PMTL in many cases. Rather, it would be more accurate to consider PMTL, in part, as an advanced symptom of dental disease when interpreting missing teeth in the bioar- chaeological record. *Correspondence to: Laura E. Cirillo Department of Anthropology University of Nevada, Reno E-mail: misslauracirillo@gmail.com This paper was the recipient of the Albert A. Dahlberg prize awarded by the Dental Anthro- pology Association in 2021. Keywords: Postmortem tooth loss, Dental pathology, Dental disease 22 Dental Anthropology 2022 │ Volume 35│ Issue 01 tissues that make up the dentition. Teeth are en- closed in some of the most fragile bone, the alveo- lar sockets of the maxilla and mandible. That brittle enclosure is susceptible to much more damage in archaeological contexts than the teeth themselves, and results in significant tooth loss after death. This article explores the potential influence of missing teeth on the analysis of skeletal samples utilizing a statistical examination of patterns in dental pathology to infer what information may have been lost from teeth missing postmortem. We will focus on patterns found in various samples that exhibit missing teeth to potentially correct the underrepresentation of oral disease prevalence and will propose steps to correct possible biases from data loss. Taphonomy of Tooth Loss Tooth loss after death can occur through tissue loss during natural processes of decomposition. The conical shape of roots, especially of anterior teeth, makes teeth susceptible to coming loose from their sockets (Oliveira, Melani, Antunes, Freitas, and Galvão, 2000). The burial environment also affects decomposition of the soft and hard tissues and in- fluences postmortem tooth loss (PMTL). In addi- tion to tissue shrinkage and decomposition during skeletonization, handling of the remains during excavation, examination, transport, and storage can contribute to the dislodging and loss of teeth (Ðurić, Rakočević, and Tuller, 2004; Oliveira et al., 2000). A common storage method for crania, for example, is to rest them on their mandibles for sta- bility, which may damage maxillary teeth (Oliveira et al., 2000). The postmortem interval, root morphology and number, and excavation methods all influence the rate of PMTL (Tibbett and Carter, 2008). Recent bioarchaeological literature emphasizes the need for careful excavation to ensure the complete re- covery of the dentition. Because skulls are often recovered with teeth missing, it is important to maximize tooth recovery through careful excava- tion methods. Loose teeth that are outside of ex- pected anatomical position may not be recognized during excavation, especially if burial context is not carefully examined (Ðurić et al., 2004). The lack of standard excavation methods has affected the way human remains are analyzed in both bioar- chaeological and forensic contexts (Evis, Hanson, and Cheetham, 2016; Haglund, 1997). Recent re- search suggests that a stratigraphic excavation method results in more evidence recovery than an arbitrary level method, especially in small element recovery rates and with fewer bones categorized as unassociated (Evis et al., 2016, Tuller and Ðurić, 2006). The recovery of disarticulated material, such as dental remains, is crucial for constructing bio- logical profiles and paleoepidemiology research (Tuller and Ðurić, 2006). Pathology of Tooth Loss Although PMTL in bioarchaeological contexts is often due to carelessness during excavation, the amount of effective soft tissue holding a tooth in its anatomical position also is an important factor to consider (Ðurić et al., 2004). Periodontal disease influences the integrity of the periodontal ligament that helps anchors the cementum to the alveolar bone and the gingiva, and therefore contributes to the potential for teeth to be easily dislodged post- mortem (Ðurić et al., 2004; Meller, Urzua, Moncada, and von Ohle, 2009). Oral disease can be introduced through several different pathways and can affect both the soft and hard tissues of the oral cavity. Teeth are at risk for loss through infection of the adjacent tissues or due to trauma to the enamel structure. The three main pathological conditions of interest are dental caries (carious lesions in the tooth), periapical lesions (lytic lesions in the alveolar bone), or occlusal tooth wear (loss of tooth enamel). These pathological conditions threaten the integrity of the tissues in- volved, therefore compromising the tooth as a unit. The most significant outcome, no matter the patho- genesis, is loss of the overall tooth. Once a tooth is lost, both the soft tissue and the surrounding alveolar bone begin to heal. Within eight weeks of tooth loss, most of the socket is filled with remodeled bone (Larjava, 2012). This remodeling reaches the alveolar crest within three to four months (Shiroma, Terrado-Naguinlin and Zuerlein, 2019), and continues for around six months, with variation based on the location and presence of neighboring teeth (Larjava, 2012). But the successful healing of a single tooth socket does not spare the rest of the oral cavity from a similar fate. Typically, the interaction of the environment and the tissues of the mouth are not confined to one tooth alone; oral pathological conditions often have multiple causes, and more than one tooth may be affected by the same disease process. As the dental tissues respond and react at different rates, moving beyond an individual tooth and con- sidering the implications of oral pathology creates a better sense of the physical indications of an indi- vidual’s overall health. From there, population lev- el analysis provides perspective on overall disease prevalence in a past community. Although rarely recorded beyond an inventory, 23 Dental Anthropology 2022 │ Volume 35│ Issue 01 tooth loss is often included in the larger interpreta- tions of prehistoric dentitions (Costa, 1980; Lukacs, 2007). The loss of data from absent teeth is one of the most prevalent concerns in the bioarchaeology dental disease literature (Cucina and Tiesler, 2003). Potential underestimation of dental caries rates has been acknowledged in calculations of disease prev- alence in samples with high rates of missing teeth (Lukacs, 1995; Littleton and Frolich, 1993). Analyz- ing when tooth loss took place (e.g., antemortem or postmortem), and the underlying factors that led to loss of that tooth, can be difficult to determine. Es- tablishing when tooth loss occurred depends on the remaining alveolar bone and the degree to which it has remodeled. Antemortem tooth loss (AMTL) of the perma- nent dentition is associated with advanced stages of dental disease. Antemortem tooth loss has sever- al potential causes: caries, pulpitis, or periodontitis resulting from infection of the tooth and the sur- rounding tissues, or trauma (Costa, 1980; Hillson, 1996; Indriati and Buikstra, 2001; Larsen, 1995). The bias in data collection that results from AMTL has long been acknowledged, primarily in the context of studying rates of dental caries, but only a few researchers have attempted to correct for this loss in data (Hardwick, 1960; Brothwell, 1963; Powell, 1985; Kelley, Levesque, and Weidl, 1991; Lukacs, 1995; Gagnon and Wiesen, 2013). Most successful- ly, Lukacs proposed the “Caries Correction Fac- tor,” which derives from the prevalence of pulp exposure due to carious lesions versus attrition observed in the sample. By creating a sample- or population-specific equation for calibrating caries rates, the correction factor considers the relation- ship between carious lesions and AMTL (Lukacs, 1995). This focus on the effects and interpretations of AMTL has led to increased incorporation of tooth loss data in dental inventories and oral dis- ease assessments (Nelson, Lukacs, and Yule, 1999; Cucina and Tiesler, 2003). Postmortem tooth loss, on the other hand, has been much more ignored in the bioarchaeological literature and often treated as missing data. Alt- hough it is commonly acknowledged as a data col- lection bias, it is rarely addressed outside of the need for careful excavation when exhuming hu- man remains (Tuller et al., 2004). As discussed below, a tooth lost either ante- mortem or postmortem is often an indication of the subtle changes in the surrounding tissues, and con- sequently the disease of the overall oral cavity, ra- ther than solely an unfortunate consequence of taphonomic processes. This study examines sam- ples that exhibit different patterns of PMTL and explores how these patterns influence the un- derrepresentation of oral disease prevalence and how to correct for this bias. Materials and Methods The dentition of 771 individuals was inventoried and each tooth or empty tooth socket was assessed for wear and pathological conditions. The methods follow Bartelink’s (2006) modification of the scor- ing system presented in Buikstra and Ubelaker’s (1994) Standards for Data Collection from Human Skel- etal Remains, to provide consistency between the new data collected and Bartelink’s 2004-2012 col- lection of dental data for the final pooled sample. When the tooth was present for observation, dental caries was then scored by location on the tooth and dental wear was recorded using the Smith (1984) system for anterior teeth and premolars and the Scott (1979) system for molars. When assessing teeth for dental caries, all potential lesions were probed using a dental pick and evaluated using a 10x magnification hand lens. For the context of this research, tooth condition was recorded as either present in occlusion, AMTL, or PMTL. All other cases were excluded. Neighboring tissues in this context were represented by an examination of the teeth immediately mesial and distal to the selected tooth and the alveolar bone that surrounded those teeth. In the case of the third molar, there was no distal tooth, so the second molar was its only neighbor. As dental disease is not often isolated to a single tooth, we hypothesize that the condition of the tooth was affected by the presence of carious lesions in the neighboring teeth and/or by the presence of periapical lesions in the neighboring tissues, given that periapical lesions can weaken the tissues holding a tooth within the alveolar bone. To be marked as “observable,” a tooth must have been present and in the occlusal plane, with greater than 2 mm of vertical enamel surrounding at least 50% crown circumference, eliminating overly worn and loose teeth. Subadults were re- moved from the original sample to ensure all indi- viduals had permanent dentition. Individuals with tooth loss due to potential congenital absence (judged by examining tooth positions relative to tooth types) were also excluded for ease of com- parison. The pooled data set consisted of 771 adult indi- viduals from Late Holocene (5000-200 BP) archaeo- logical sites in pre-contact California, which was created using the dental inventories and pathology assessments. The sample population was repre- sented by individuals from CA-ALA-307, -309, - 24 Dental Anthropology 2022 │ Volume 35│ Issue 01 328, and -329, sites located near the shoreline of the San Francisco Bay, the ancestral homelands of the Ohlone tribe, and from CA-SAC-06, -43, -60 and SJO-68, -142, and -154, sites located in the Central Valley, the ancestral homelands of the Plains Miwok tribe. This research used a combination of new data collected for this study and previously collected data from Bartelink’s (2006) dissertation research. All dentitions were examined at UC Berkeley’s Phoebe A. Hearst Museum of Anthro- pology, where they are currently curated. Permis- sion to collect data were provided by the muse- um’s curator and NAGPRA committee. After instances of PMTL with all observable neighboring teeth were isolated, the collection con- sisted of two groups: (i) a control group, where the primary tooth examined was present and in occlu- sion, and (ii) an experimental group, where the primary tooth examined was absent postmortem. Teeth were pooled from right and left sides of the mandible and maxilla. Tooth counts of the total sample are presented in Table 1. Results The data analysis first considers whether instances of PMTL were more often associated with sur- rounding teeth or neighboring tissues that had al- ready experienced AMTL. The presence or absence of the neighboring teeth was compared between each primary tooth that was present and in occlu- sion, and those recorded as PMTL. After organiz- ing by tooth type (Table 2), the percentage of pri- mary teeth with both neighboring teeth present was lower in every tooth group when the primary tooth being examined was lost postmortem. The smallest difference was a 2.3% decrease in third molars, and the largest difference was a 25.4% de- crease in the fourth premolar. The average differ- ence between having all neighboring teeth present between control teeth and PMTL teeth was a 15.3% decrease when all tooth types were considered. The visual representation (Figure 1) shows that the percent difference was especially high for posterior teeth. The average difference between anterior tooth types was a 12.3% decrease. The average in posterior teeth was a 17.1% decrease (20.8% when third molars were excluded). When third molars were excluded for not meet- ing the criteria of having two neighboring teeth, instances of having one neighboring tooth present and the other absent were most often seen in the posterior teeth. In this analysis, greater than 20% of PMTL affected second molars, first molars, fourth premolars, and third premolars had one present and one absent neighbor. Although the change in percentage of teeth with two neighbors present was not as great between control and PMTL inci- sors, all anterior teeth showed a consistent decline in percent of present neighbors in each experi- mental group. Rather than a similarly high preva- lence of the one present and one absent neighbor alternative, as was seen in posterior teeth, all the anterior teeth were more affected by AMTL on both sides. Dental Caries and Neighboring Tissues To understand the effects of specific dental patho- logical conditions on the prevalence of PMTL, an analysis of the neighboring tissues was also con- ducted to see how carious lesions and periapical lesions are associated with compromised sur- rounding tissues and overall tooth loss (Table 3). In this analysis, the control tooth was present without evidence of carious lesions, and the experimental tooth was again one in the same position that was lost postmortem. Although all analyses showed that it is rare to have both neighboring teeth affected by the same pathological condition, in the case of carious le- sions, there were two cases seen on canine teeth. Both control and experimental groups showed few differences when both neighboring teeth were pre- sent. The smallest difference in percent of all neigh- boring teeth with no caries was 0.8% in second mo- lars, while the greatest difference was only 7.8% in first molars. Neighboring teeth of experimental groups for third premolars, canines, and both inci- sors all displayed no caries. Consistent with the literature on dental caries, posterior teeth were more affected than anterior teeth. Most teeth showed a slight decrease in caries prevalence in the surrounding tissues in the experi- TOOTH # CONTROL (PRESENT) # EXPERIMENTAL (PMTL) M3 930 166 M2 815 20 M1 924 12 P4 854 118 P3 765 78 C 650 74 I2 522 113 I1 476 73 Table 1. Research sample size by tooth and condition. 25 Dental Anthropology 2022 │ Volume 35│ Issue 01 Tooth Condition All neighboring teeth present (%) 1 tooth present, 1 tooth AMTL (%) All neighboring teeth AMTL (%) M3 Present 93.9 4.9 n/a PMTL 91.6 8.4 n/a M2 Present 92.4 6.1 1.5 PMTL 75.0 25.0 0.0 M1 Present 94.4 3.4 0.0 PMTL 75.0 25.0 0.0 P4 Present 96.6 3.3 0.5 PMTL 71.2 21.2 5.0 P3 Present 97.8 2.2 0.0 PMTL 76.9 21.8 1.3 C Present 97.5 1.8 0.6 PMTL 78.4 9.5 12.2 I2 Present 97.7 2.3 0.0 PMTL 87.6 7.1 5.3 I1 Present 99.6 0.4 0.0 PMTL 91.8 4.1 4.1 Table 2. Prevalence of all three neighboring tissue categories. Figure 1. Visual comparison of the prevalence of neighboring tissue categories. 26 Dental Anthropology 2022 │ Volume 35│ Issue 01 mental group. Control groups for third molars, second molars, third premolars, canines, and all incisors each had a higher percent of neighbors with carious lesions than their experimental coun- terparts. When isolated to show teeth that had one neighboring tooth present and one with AMTL, carious lesions were only observed in posterior teeth, consistent with existing literature. Caries rates were again higher in the control group, with a large increase from 80% present second molars with the observable neighbor displaying no caries, to 100% in the neighbors of the PMTL second mo- lars. There was no caries-focused analysis of the difference between present and PMTL teeth with both neighbors absent because, unlike periapical lesions that affect tissue other than on the actual tooth, neighboring teeth necessarily must be pre- sent to observe dental caries lesions. Periapical Lesions and Neighboring Tissues Following the framework of the caries-focused ex- amination of neighboring teeth, the neighboring tissues (both tooth and alveolar socket) were exam- ined for periapical lesions. Table 4 shows a com- parison of the prevalence of periapical lesions when both neighboring teeth were present. The experimental group in every tooth group had more periapical lesions in the neighboring tissues than the control group, except for first molars which had a 1.0% increase in prevalence of no periapical lesions when the primary tooth was recorded as PMTL. The smallest change was a 0.4% decrease in central incisors, and the greatest change was the 18.4% decrease in lateral incisors. Generally, periapical lesions had a greater influ- ence on non-molar teeth, which was particularly apparent for periapical lesions in teeth with one neighboring tooth present and one lost ante- mortem. Although there is no particular pattern in the posterior teeth (i.e., minimal differences ob- served between control and experimental groups), the anterior teeth and fourth premolars show a consistent decrease in unaffected neighboring tis- sue in the experimental groups (as depicted in Fig- ure 2). The difference between all control and ex- perimental groups shows an average 27% decrease (23.5-33.3%, min-max). There was not enough data to see a pattern in instances where both neighboring teeth were ab- sent, but the inability to gather enough instances of control data to provide a comparison may be tell- ing. No data showed PMTL with two AMTL neighbors, with a maximum count of nine when divided by tooth number, and with no molars fit- ting these criteria. With few exceptions, this data set failed to show a present tooth that had two AMTL neighbors. Twelve instances were recorded in second molars, four in fourth premolars, and four in canines. All other teeth had no instances of this tooth loss pattern. It is possible that some oral pathological condi- tions may not affect neighboring tissues but are more limited to the individual tooth. Given the expectation that missing neighbors will compro- mise the maintenance of the primary tooth being analyzed, especially in the case of periapical le- sions that affect tissues of the jaw as well as the tooth, it is not surprising that two missing neigh- Tooth Condition All neighboring teeth no caries (%) 1 neighboring tooth with caries (%) 2 neighboring teeth with caries (%) M3 Present 95.3 4.7 n/a PMTL 98.0 2.0 n/a M2 Present 92.6 7.4 0.0 PMTL 93.3 6.7 0.0 M1 Present 96.7 3.3 0.0 PMTL 88.9 11.1 0.0 P4 Present 97.2 2.8 0.0 PMTL 96.4 3.6 0.0 P3 Present 98.5 1.5 0.0 PMTL 100.0 0.0 0.0 C Present 98.9 0.7 0.3 PMTL 100.0 0.0 0.0 I2 Present 99.0 1.0 0.0 PMTL 100.0 0.0 0.0 I1 Present 97.9 2.1 0.0 PMTL 100.0 0.0 0.0 Table 3. Prevalence of all three dental caries inventory categories. 27 Dental Anthropology 2022 │ Volume 35│ Issue 01 Tooth Condition No PL on all neighboring teeth (%) 1 neighboring tooth with PL (%) 2 neighboring teeth with PL (%) M3 Present 96.8 3.2 n/a PMTL 94.1 5.9 n/a M2 Present 93.0 7.0 0.0 PMTL 80.0 20.0 0.0 M1 Present 99.1 0.9 0.0 PMTL 100.0 0.0 0.0 P4 Present 94.1 5.7 0.2 PMTL 84.5 14.3 1.2 P3 Present 99.9 0.1 0.0 PMTL 93.3 5.0 1.7 C Present 99.1 0.9 0.0 PMTL 98.3 1.7 0.0 I2 Present 99.2 0.8 0.0 PMTL 80.8 15.2 4.0 I1 Present 98.9 1.1 0.0 PMTL 98.5 1.5 0.0 Table 4. Prevalence of all three periapical lesion inventory categories. Figure 2. Visual comparison of the prevalence of neighboring tissues exhibiting periapical lesions when one neighboring tooth is present and one is AMTL. 28 Dental Anthropology 2022 │ Volume 35│ Issue 01 bors make the tissues less likely maintain a tooth. Thus, a large enough control group sample was unavailable for this analysis. This was the only sit- uation where the experimental sample was larger than the control sample. Affected Neighbors Because multiple dental conditions can be present in the same individual (even on the same tooth) and would not be accounted for when analysis nar- rows to focus on a single pathological condition at a time, the scope of analysis was expanded to look at the effects of neighboring tissues affected by ei- ther caries or periapical lesions. This included AMTL as an indicator of the final stage of either pathological condition, where the tooth could not be maintained in life. Figure 3 shows the percentage of neighboring teeth for each tooth position (control) and PMTL, divided by no affected neighbors, one affected neighbor, or both neighbors affected. To be catego- rized as “affected,” a tooth or its surrounding tis- sue needed to display carious lesions, periapical lesions, be absent antemortem, or any combination of the three conditions. Because third molars only have one neighboring tooth, they again could only be recorded as having one affected neighbor or no affected neighbors. There is a clear pattern in the visual comparison that PMTL is often surrounded by “affected” or tissues without any indication of disease. In the control groups, the tooth had two unaffected neighbors an average of 92.0% of the time (Table 5). By contrast, the PMTL groups had only an aver- age of only 70.5%. The smallest difference was be- tween the control and PMTL groups for third mo- lars (3.4%), while the largest difference was be- tween the control and PMTL groups for first mo- lars (36.0%). The PMTL group with the highest percent of two unaffected neighbors was the central incisor. If neighboring tissues do not contribute to its loss in a significant way, then retention of the central inci- sor is the least influenced by tissue health and tooth loss must be attributed to other, taphonomic factors. This is perhaps the most common tapho- nomic loss due to the instability of a single location at the anterior of the mouth (exposing it to maxi- mum pressure in burial and collection storage). The difference in the distribution of one affected neighbor and both neighbors affected also showed an interesting pattern. Non-molar teeth had a much greater percentage of two affected neighbors. The average of non-molar PMTL with two affected neighbors was 9.0% and varied from 6.0 to 12.0%. Although it is common to see a difference be- tween third molars and the other molars because of the unique single-neighbor quality and differ- Figure 3. Visual comparison of the prevalence of “affected” neighboring tissues. 29 Dental Anthropology 2022 │ Volume 35│ Issue 01 ences in eruption times, this is one of the only com- parisons from this research that showed variation between the second and first molars. They are typi- cally assumed to have similar physical characteris- tics that make them susceptible to caries, but also are multi-rooted teeth, providing them similar con- nective stability. Discussion The “neighboring tissues” test was designed to determine whether oral pathological conditions influenced PMTL versus solely taphonomic expla- nations. Patterns between control teeth and PMTL indicate that the integrity of the surrounding tis- sues affects the retention of a tooth after death. Alt- hough there were specific patterns associated with periapical lesions alone, the overall patterns indi- cated that dental caries does not affect the ability of the tissues to retain a tooth after death. When both neighboring teeth were present, most teeth showed a decrease in carious lesions in the surrounding teeth in the experimental (PMTL) group. This is contrary to the expectation that missing teeth would have more affected neighbors in the pres- ence of any oral pathological condition. However, the experimental group presented more periapical lesions in the neighboring tissues than the control group. Periapical lesions tended to have a greater effect on non-molar teeth. Although there were specific patterns associated with the presence of dental periapical lesions as a solitary pathological condition, dental caries lesions did not affect the ability for the tissues to retain a tooth after death. To address tissues that have been impacted vs. not impacted by disease, rather than exclude com- promising conditions by creating a false categori- zation that separates dental caries and periapical lesions, the data were lumped into an “affected tissues” test. This clear difference between control and experimental groups supports the research hypothesis that affected tissues are correlated with prevalence of PMTL. Overall, the analyses conducted indicate that PMTL is often a consequence of pathological con- ditions rather than solely due to taphonomic dam- age. Thus, it should be possible to adjust data relat- ed to PMTL in the same way AMTL is corrected to generate more reliable caries rates. Using indica- tors of disease in the surrounding tissues, the pres- ence of a pathological condition in the missing tooth can potentially be inferred, adjusting preva- lence rates in a skeletal sample. As with the caries correction factors for AMTL, a sliding scale or pop- ulation-specific method is needed. This would need to be calculated based on the integrity of the visible tissues and can only be accomplished if PMTL is considered a consequence of pathology, rather than simply the result of postmortem dam- age to the alveolar bone. More elaborate and pre- cise observations need to be incorporated into the Tooth Condition All neighboring teeth unaffected (%) 1 neighboring tooth affected (%) 2 neighboring teeth affected (%) M3 Present 88.3 11.7 n/a PMTL 84.9 15.1 n/a M2 Present 78.8 17.7 3.5 PMTL 55.0 45.0 0.0 M1 Present 94.4 5.6 0.0 PMTL 55.0 41.7 0.0 P4 Present 88.8 10.1 1.2 PMTL 62.7 28.0 9.3 P3 Present 96.2 3.5 0.0 PMTL 69.2 21.8 9.0 C Present 95.5 3.5 1.0 PMTL 75.7 12.2 12.2 I2 Present 96.6 3.1 0.3 PMTL 69.9 21.2 8.8 I1 Present 97.1 2.1 0.8 PMTL 88.7 5.6 5.6 Table 5. Prevalence of all three “affected” inventory categories 30 Dental Anthropology 2022 │ Volume 35│ Issue 01 inventory methodology. Current inventory recommendations only con- sist of practicing extra care while analyzing the fragile alveolar bone and recording missing teeth as “absent, without alveolar bone remodeling, postmortem tooth loss”. There is no current meth- od for recording teeth that are loose and replaced in their crypt, other than recording them as pre- sent, which does not distinguish them from teeth that are maintained in the crypt by supporting tis- sues. A new category should be added to the in- ventory methods to reflect this difference when inventorying teeth. If teeth are “absent through postmortem tooth loss” or “present but loose/removable from crypt without force”, observations can be made to exam- ine the empty crypt for signs of pathology in the tissue. A closer look during analysis using clues of the surrounding tissue may help indicate the health of the missing teeth, given our improved understanding of how pathological conditions spe- cifically affect tooth retention. Collecting this data will permit a wider range of calculations of dental pathology prevalence for data sets. REFERENCES Bartelink, E. (2006). Resource intensification in pre- contact Central California: A bioarchaeological per- spective on diet and health patterns among hunter- gatherers from the Lower Sacramento Valley and San Francisco Bay (doctoral dissertation). Texas A&M University, College Station, Texas. Brothwell, D. (1963). The macroscopic dental pa- thology of some earlier human populations. In: Dental Anthropology. D. Brothwell (Ed.). Lon- don: Pergamon Press. Buikstra, J.E, & Ubelaker, D.H. (1994). Standards for Data Collection from Human Skeletal Remains. Proceedings of a Seminar at the Field Museum of Natural History (Arkansas Archeological Report Research Series). Costa, R. L. (1980). Age, sex, and antemortem loss of teeth in prehistoric Eskimo aamples from Point Hope and Kodiak Island, Alaska. Ameri- can Journal of Physical Anthropology, 53, 579-587. Cucina, A., & Tiesler, V. (2008). Dental caries and antemortem tooth loss in the Northern Peten Area, Mexico: A biocultural perspective on social status differences among the classic Ma- ya. American Journal of Physical Anthropology, 122(1), 1-10. Ðurić, M., Rakočević, Z., & Tuller, H. (2004). Fac- tors affecting postmortem tooth loss. Journal of Forensic Science, 49(6). Erdal, Y. S., & Duyar, I. (1999). A new correction procedure for calibrating dental caries frequen- cy. American Journal of Physical Anthropology, 108(2), 237-240. Evis, L.H., Hanson, I., & Cheetham, P.N. (2016). An experimental study of two grave excavation methods: Arbitrary Level Excavation and Strat- igraphic Excavation. STAR: Science & Technolo- gy of Archaeological Research, 2(2), 177-191. Gagnon, C.M. & Wiesen, C. (2013). Using general estimating equations to analyze oral health in the Moche Valley of Peru. International Journal of Osteoarchaeology, 23, 557-572. Haglund, W.D. (1997). Scattered skeletal human remains: search strategy considerations for locating missing teeth. In: Forensic Taphonomy: The Postmortem Fate of Human Remains. Boca Raton: CRC Press. Hardwick, J.L. (1960). The incidence and distribu- tion of caries throughout the ages in relation to the Englishman’s diet. British Dental Journal, 108, 9. Hillson, S. (1996). Dental Anthropology. Cambridge University Press: Cambridge, UK. Hillson, S. (2005). Teeth (Second Edition). Cam- bridge University Press: Cambridge. Indriati, E. & Buikstra, J.E. (2001). Cocoa chewing in prehistoric Coastal Peru: Dental evidence. American Journal of Physical Anthropology, 114 (3), 242-257. Kelley, M. A., Levesque, D.R., & Weidl, E. (1991). Contrasting patterns of dental disease in five early northern Chilean groups. In: Advances in Dental Anthropology. M. A. Kelley & C. S. Larsen (Eds.). New York: Wiley-Liss. Konig, K. G. (2000). Diet and oral health. Interna- tional Dental Journal, 50(3), 162-174. Larjava, H. (2012). Oral Wound Healing: Cell Biology and Clinical Management. Wiley-Blackwell: West Sussex. Larsen, C.S. (1995). Biological changes in human populations with agriculture. Annual Review of Anthropology, 24, 185-213. Littleton, J. & Frolich B. (1993). Fish-eaters and farmers: Dental pathology in the Arabian Gulf. American Journal of Physical Anthropology, 92, 427-447. Luby, E.M. (2004). Shell mounds and mortuary behavior in the San Francisco Bay Area. North American Archaeologist, 25(1), 1-33. Lukacs, J.R. (1995). The ‘Caries Correction Factor’: A new method of calibrating dental caries rates to compensate for antemortem loss of teeth. International Journal of Osteoarchaeology, 5, 151- 31 Dental Anthropology 2022 │ Volume 35│ Issue 01 156. Lukacs, J. R. (2007). Dental trauma and ante- mortem tooth loss in prehistoric Canary Is- landers: prevalence and contributing factors. International Journal of Osteoarchaeology, 17(2), 157-173. Moynihan, P. (2005). The interrelationship between diet and oral health. Proceedings of the Nutrition Society, 64, 571-580. Meller, C., Urzua, I., Moncada, G. & von Ohle, C. (2009). Prevalence of oral pathologic findings in an ancient pre-Columbian archaeologic site in the Atacama Desert. Oral Diseases, 15, 287- 294. Nelson, G.C., Lukacs, J.R., & Yule, P. (1999). Dates, caries, and early tooth loss during the Iron Age of Oman. American Journal of Physical Anthro- pology, 108(3), 333-343. Oliveira, R.N., Melani, R.F.N., Antunes, J.L.F., Freitas, E.R., & Galvão, L.C.C. (2000). Postmor- tem tooth loss in human identification process- es. The Journal of Forensic Odonto-Somatology, 18 (2), 32-36. Pilloud, M. A. & Fancher, J.P. (2019). Outlining a definition of oral health within the study of human skeletal remains. Dental Anthropology, 32(2), 3-11. Powell, M. L. (1985). The analysis of dental wear and caries for dietary reconstruction. In R. I. Gilbert & J. H. Mielke (Eds.) The Analysis of Prehistoric Diets (pp. 307-338). Orlando: Aca- demic Press. Scott, G.R. & Turner II, C.G. (1988). Dental anthro- pology. Annual Review of Anthropology, 17, 99- 126. Scott, E.C. (1979). Dental wear scoring technique. American Journal of Physical Anthropology, 51(2), 213-217. Seo, B.M., Miura, M., Gronthos, S., Bartold, P.M., Batouli, S., Brahim, J., Young, M., Gehron Rob- ey, P., Wang, C., & Shi, S. (2004). Investigation of multipotent postnatal stem cells from hu- man periodontal ligament. Lancet, 364, 149-55. Shiroma, C.Y., Terrado-Naguinlin, P.M., & Zuer- lein, C.L. (2019). Healing alveolar sockets in skeletonized remains: A report on cases from one month to twelve months post-extraction. Forensic Science International, 301, e38-e43. Smith, B.G., & Knight, J.K. (1984). An index for measuring the wear of teeth. British Dental Journal, 156(12), 435-8. Tibbett, M. & Carter, D.O. (2008). Soil Analysis in Forensic Taphonomy: Chemical and Biological Ef- fects of Buried Human Remains. CRC Press: Boca Raton. Tuller, H. & Ðurić, M. (2006). Keeping the pieces together: Comparison of mass grave excava- tion methodology. Forensic Science International, 156, 192-200. Tuller, H., Durio, M., & Rakočević, Z. (2004). Fac- tors affecting postmortem tooth loss. Journal of Forensic Sciences, 49(6), 1313-1318.