Chronology of Linear Enamel Hypoplasia Formation in the Krapina Neanderthals

During childhood, systemic physiological stresses such as illness, disease, and malnutrition can disrupt the growth of dental enamel. These disruptions are often recorded in the form of linear enamel hypoplasia (LEH). Many researchers have analyzed the frequency and timing of LEH formation in Neanderthal populations as they relate to ideas about Neanderthal living conditions, nutrition, and foraging efficiency. Previous age estimates for Neanderthal LEH were largely based upon modern human dental growth standards. However, recent studies provide a more complete picture of Neanderthal tooth formation. We use data from these studies to create enamel growth charts for four Neanderthal anterior tooth types (upper central and lateral incisors, upper and lower canines) analogous to those created for modern humans by Reid and Dean (2000). The Neanderthal charts differ from those of modern humans especially in initiation ages and in the duration of enamel formation within equivalent divisions of crown height. Based on these new charts, we estimate ages at formation for a series of Krapina Neanderthal defects. We also compare estimated ages at defect formation in the Krapina sample with estimated ages of defect formation in a sample of modern humans from Point Hope, Alaska. The median ages at defect formation across different anterior tooth types range from 2.3–2.5 (based on a seven-day perikymata periodicity) and 2.6–2.8 years (based on an eight-day perikymata periodicity), suggesting that Neanderthals experienced physiological stress earlier in life than indicated by previous estimates that were derived from modern human standards. By contrast, median ages at defect formation in the Point Hope sample are later than those of the Krapina Neanderthals, which may result from differences in crown growth geometry between Neanderthals and modern humans, differences between the two populations in the ages at which they experienced episodes of stress, or both. E hypoplasias are developmental defects that reflect periods of disrupted enamel growth most commonly caused by periods of malnutrition, undernutrition, or illness (Goodman and Rose 1990; Hillson 2014). Because these defects are markers of such systemic physiological stresses, they have figured promineny in the Neanderthal literature (Brennan 1991; Guatelli-Steinberg et al. 2004; Hutchinson et al. 1997; Molnar and Molnar 1985; Ogilvie et al. 1989; Skinner 1996). Enamel hypoplasias in Neanderthals are of particular interest because they provide evidence that can yield insight into whether Neanderthals lived under conditions of nutritional stress (Jelinek 1994) and/or were inefficient foragers (Binford 1989; Soffer, 1994; Trinkaus 1986, 1989; but see Sorensen and Leonard 2001). With one exception (Brennan 1991), studies of enamel hypoplasias in Neanderthals focus on the Krapina remains PaleoAnthropology 2014: 431−445. © 2014 PaleoAnthropology Society. All rights reserved. ISSN 1545-0031 doi:10.4207/PA.2014.ART84 DEBBIE GUATELLI-STEINBERG Department of Anthropology and Department of Evolution, Ecology and Organismal, Biology, 4034 Smith Laboratory, 174 W. 18th Avenue, Ohio State University, Columbus, OH 43210, USA; [email protected] ASHLEY STINESPRING-HARRIS Department of Anthropology, 109 Davenport Hall, 607 S. Matthews Avenue, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USA; [email protected] DONALD J. REID Department of Anthropology, 2110 G. Street, George Washington University, Washington D.C. 20052, USA; [email protected] CLARK SPENCER LARSEN Department of Anthropology and Department of Evolution, Ecology and Organismal, Biology, 4034 Smith Laboratory, 174 W. 18th Avenue, Ohio State University, Columbus, OH 43210, USA; [email protected] DALE L. HUTCHINSON Department of Anthropology, 301 Alumni Building, CB # 3115, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, USA; [email protected] TANYA M. SMITH Department of Human Evolutionary Biology, 11 Divinity Avenue, Harvard University, Cambridge, MA 02138, USA; [email protected] 432 • PaleoAnthropology 2014 drawn with caution, this finding suggests that Krapina Neanderthals did not experience stress episodes of longer duration than did the Inupiaq, especially if Neanderthals had slightly lower periodicities than modern humans (Smith et al. 2010). The comparison of stress episode duration between the Krapina Neanderthals and the Point Hope Inupiaq is meaningful because both groups were foragers, with some similarities in the environments they inhabited. The Krapina Neanderthals appear to have inhabited a deciduous woodland environment (Fiorenza et al. 2011) at a time of “rapidly changing climates and oscillating landscapes” (Hutchinson et al. 1997: 912), while the Point Hope Inupiaq occupied a marginal arctic habitat. These are clearly not identical habitats, but each presents environmental challenges to a foraging way of life. If, as has been claimed (Binford 1989; Soffer 1994; Trinkaus 1986, 1989; but see Sorensen and Leonard 2001), Neanderthals were inefficient foragers relative to modern humans, then they might have been expected to have recorded in their enamel evidence of more prolonged stress episodes than did the Inupiaq. The perikymata evidence, however, does not support this view. Here, we extend the comparison between LEH defects in the Krapina Neanderthals and the Point Hope Inupiaq to examine the chronology of stress episode occurrence. We also investigate whether there are differences between the two groups in the ages at which stress episodes, as represented by LEHs, occurred. To accomplish this comparison, we make use of recent studies on Neanderthal dental development (Guatelli-Steinberg et al. 2005, 2007; Smith et al. 2007b, 2010) to provide the most accurate estimates of LEH defect formation in Neanderthals currently possible. In so doing, we create charts for aging LEH defects in Neanderthals, analogous to those created for modern humans by Reid and Dean (2000) for four Neanderthal anterior tooth types. MATERIALS AND METHODS ESTABLISHING GROWTH CHARTS FOR NEANDERTHAL AND INUIT ANTERIOR TOOTH CROWNS Reid and Dean (2000) created a series of enamel growth charts based on enamel histological sections made from 115 unworn anterior teeth (routinely extracted from dental patients living in the United Kingdom). Data on the ages at which mineralization began for each tooth type were taken from an earlier publication (Reid et al. 1998). These ages were added to the average time taken to form cuspal enamel for each tooth type, giving the age at which cuspal enamel formation was completed. Next, the authors calculated the subsequent lateral enamel formation time, which includes the entire crown height. To calculate lateral enamel formation time, the authors counted the internal striae of Retzius, which crop out on the enamel surface as perikymata, and form with a regular periodicity of six to 12 days in different individuals (Reid and Dean 2006). To determine the periodicity of striae, daily increments known as cross (Hutchinson et al. 1997; Molnar and Molnar 1985) or include a large proportion of them (Guatelli-Steinberg et al. 2004; Ogilvie et al. 1989; Skinner 1996) because these fossils constitute the largest number of Neanderthal individuals from a single site (Radovčić et al. 1988). By incorporating or focusing on the Krapina dental remains, these enamel hypoplasia studies provide a glimpse into population-level stress in Neanderthals, at least in one especially well-studied setting in Central Europe. While these studies generally find high frequencies of enamel hypoplasia in the Krapina Neanderthals (using various measures and analyzing different types of enamel hypoplasia), these frequencies do not appear to differ from the high frequencies of enamel hypoplasias observed among some modern foraging groups (Guatelli-Steinberg et al. 2004; Hutchinson et al. 1997). Thus, while the enamel hypoplasia evidence is consistent with the hypothesis that the Krapina Neanderthals experienced nutritional stress, it does not lend support to the contention that they were more nutritionally stressed than were (or are) some foraging populations of anatomically modern humans. Linear enamel hypoplasia (LEH) is the most common form of the different types of enamel hypoplasia (Hillson and Bond 1997). LEHs appear as horizontal lines, grooves, furrows, or linear arrays of pits on the enamel surface (FDI DDE Index 1982, 1992; Goodman and Rose 1990; Hillson and Bond 1997). Unlike other enamel hypoplasias, LEHs are informative about the duration or chronology of the stress episodes they represent (Hillson and Bond 1997). They can yield this information because, unlike other forms of enamel hypoplasia, LEHs are directly associated with enamel growth increments that manifest on the enamel surface as perikymata. Within the teeth of an individual, these layers take a constant number of days to form, although across modern human individuals, perikymata represent a range of six to 12 days of growth (Reid and Dean 2006). A study of 11 Neanderthals revealed a range of six to nine days across individuals (Smith et al. 2010). With the exception of linear arrays of pits, linear enamel hypoplasias are composed of one to several perikymata (Hillson and Bond 1997). Consequently, the time span represented by LEH defects, as well as the ages at which they were formed, can be determined. Unfortunately, the number of days represented by perikymata (their periodicity) and the initiation at which crown formation began are not apparent from the tooth surface. Thus, barring physical or virtual sectioning (Tafforeau and Smith 2008), the duration and timing of the stress episodes that LEH defects represent are usually estimated from previously published data. These estimates, however, provide greater detail about physiological stress experience than do LEH frequency data alone. For example, Guatelli-Steinberg et al. (2004) found LEH frequencies to be similar in their Neanderthal and Inupiaq samples. However, they also found that the average number of perikymata in seven Inupiaq defects (13.4 perikymata) was statistically significantly greater than that found for 15 Krapina defects (7.3 perikymata). Although these sample sizes are small such that inferences must be Linear Enamel Hypoplasia Formation in Krapina Neanderthals• 433 made by visually extending the converging sides of the worn crown. For incisors, crown height estimates were made by comparing the morphology of unworn crowns to worn crowns. Finally, two of the Krapina Neanderthal teeth included in this study (Krapina 91 and 93) are incomplete crowns on which the majority of the crown appears to have formed. To estimate the completed crown height for these teeth, the lines of curvature on mesial and distal sides of crown near the cervix were visually extended. For Krapina 91 and 93, we estimated that 89% of the crown height had been completed. Some previous studies of LEH prevalence have limited analyses only to those teeth on which perikymata could be at least partially observed (Guatelli-Steinberg 2003, 2004; Guatelli-Steinberg et al. 2004). Such studies have done so because abrasion great enough to remove perikymata from the enamel surface might also remove minor LEH defects, affecting estimates of LEH prevalence. However, in this study the aim is to age defects, not assess prevalence. Thus, here we included all anterior permanent teeth which were estimated to be 80% or more complete, regardless of whether perikymata were observable on them. LEH defects were identified under conditions of diffuse lighting with a second light source oriented obliquely to the specimen (Goodman and Rose 1990; Lukacs 1989). A 10x hand lens aided in identifying defects. The first author examined the original Point Hope teeth and coated replicas (see below) of the Krapina teeth. The lower limit of defects identified in this study were lines or grooves that appeared to be larger than adjacent perikymata grooves under 10x magnification. The upper limit was prominent lines or grooves of varying depth and width that were clearly visible without magnification. Defects were measured using Mitutoyo digital calipers from the CEJ to the middle of each defect along the midline of the tooth. To assess measurement error, 25 defects were measured three times consecutively. The average difference between the first and third measurement is 0.18mm (ranging from 0.02mm to 0.47mm), representing an average measurement error of 4%. Subjectivity in where to place the tips of the calipers, particularly in judging the “middle” of a wide defect, seems likely to be the primary source of measurement error. To assign an estimated age at defect formation using the enamel growth charts, distances from the CEJ were divided by actual crown heights in the case of unerupted unworn crowns or by estimated completed crown heights in the case of worn (or the two incomplete) crowns. The result of this division (i.e., the quotient) multiplied by 10 gives the decile in which the defect lies. That decile was located on the enamel growth chart and an age at formation was assigned. Most often defects do not fall exactly at the borders of a decile, such that interpolations between the ages at which a decile is completed must be made. To do so, Martin et al. (2008) used a nonlinear interpolation (to account for the non-linear pattern of tooth growth) as well as a linear interpolation, but these interpolations produced similar estimated ages at defect formation. Thus, here we follow Martin et al. (2008) in using a linear interpolation. Hillson (1996, 2014) points out that if a systemic, rathstriations were counted between striae. While periodicities vary across individuals, they are constant for all of the teeth of a single individual (FitzGerald 1998). Thus, by counting all of the striae of Retzius present in lateral enamel and multiplying by their periodicity, lateral enamel formation time was determined. To facilitate the aging of LEH defects, Reid and Dean (2000) divided the crown height into deciles, and used the mean number of striae of Retzius to calculate the mean age at which each decile of crown height was completed. These charts provide a model for establishing enamel growth charts for Neanderthal and Point Hope Inupiaq anterior tooth crowns in the present study. We used initiation and cuspal enamel formation time estimates from Reid and Dean (2000) to create charts for aging Point Hope defects. However, for the lateral enamel, we used previously published perikymata counts per decile from the Point Hope teeth (Guatelli-Steinberg et al. 2007). Because we did not have enamel sections of these teeth, we used periodicities of eight or nine days, the two most common values for periodicities in a variety of modern human groups (Smith et al. 2007a). Thus, these lateral enamel formation times are estimates based on central tendencies in the data. To create charts for aging Neanderthal defects, initiation and cuspal enamel formation times were taken from data in Smith et al. (2007b, 2010). Methods for determining initiation ages and cuspal enamel formation times follow established conventions and are detailed in these papers. For the lateral enamel, we used previously published perikymata counts per decile from Neanderthal teeth (Guatelli-Steinberg et al. 2007). Again, without enamel sections of these particular teeth, we used the two most common values of seven and eight days for Neanderthals (Smith et al. 2010). Because periodicities can range from six to nine days in Neanderthals (Smith et al. 2010) and six to 12 days in modern humans (Reid and Dean 2006), the lateral enamel formation times given in this paper are estimates.