Human Growth and Stature
Rebecca Gowland & Lauren Walther
Introduction
Stature is the most ancient form of biometric data collected from human populations and continues to be one of the most frequently recorded physiological parameters today. While skin wrinkles, hair is lost, and weight fluctuates; height is considered to be a stable identifying feature of adults. In actuality, stature does decrease slightly in older age and can even fluctuate by 1.5–2 centimeters throughout the course of a single day. Between waking and sleeping, compression of the soft tissues occurs during weight-bearing activities, thus reducing height. By contrast, astronauts enduring prolonged periods of weightlessness may ‘grow’ in height by as much as four to six centimeters because of abnormal expansion of the inter-vertebral discs.1 After returning to earth, their stature undergoes a period of readjustment, during which they are at risk of back injury. For the rest of us mere earth dwellers, fluctuations in stature are relatively minor, and height is considered a robust descriptor of individuals and populations.
Despite this, final adult stature is a product of the interplay between an individual’s genetics and a range of environmental and social factors impacting on the body during the growth period.2 As a consequence of extrinsic influences, a substantial degree of intrapopulation variation can occur, even within genetically homogenous populations.3 The sensitivity of stature to life circumstances means that it can be usefully harnessed to investigate the impact of a variety of environmental and social variables on the health of populations.4 Steckel provides a valuable synthesis of the range of applications of stature data, including studies of colonialism, migration, slavery, infectious disease and occupation.5
In today’s industrialized world “taller populations are generally richer populations.”6 Studies of both ancient and modern groups demonstrate that taller individuals tend to have longer average life expectancies than their shorter counterparts.7 This is because short adult stature is associated with adverse conditions during childhood, which can prevent an individual from achieving their genetic height potential. However, growth stunting during childhood can be masked by the phenomenon of “catch-up growth”: an accelerated period of growth following an earlier episode of stasis.8 Nonetheless, in order for catch-up growth to occur, the child requires improved nutritional and environmental conditions; further, there has to be something remaining of the potential growth-period.
The 1958 British Birth Cohort Study found that the height of children at age seven was a useful predictor of future employment status because growth provides a sensitive indicator of social conditions. The chances of unemployment in the shortest fifth of children were found to be three time greater than the tallest fifth. In this study, height at seven years was viewed to be reliably correlated with poor socioeconomic and psychosocial environment.9 In a similar vein, historical accounts from nineteenth-century England frequently attest to stunted growth and the lamentable physiological condition of child factory workers, with social commentators such as Engels warning that child labor would result in a “race of pygmies.”10 Public health reformers of the period, such as Edwin Chadwick measured the heights of children and adults, but ultimately determined that the unhygienic urban environment, rather than factory work was responsible for the observed growth deficits in children and small adult stature.11
The relationship between adult stature and health is contingent upon the growth journey rather than the endpoint. This is well-illustrated by Barker and colleagues’ study of the Helsinki birth cohort, which demonstrated that boys who were tall when they entered school (indicative of adequate nutrition and environment) had a longer life span.12 However, those boys who were tall as a consequence of rapid catch-up growth after a period of stunting had shorter life spans. In this example, adult stature was comparable, but the hidden consequences of earlier life stressors were expressed in terms of elevated frailty in later life.
Studies of adult stature in past populations have traditionally drawn upon historical records. Such data, however, is heavily skewed towards more recent populations, usually of eighteenth–nineteenth century date or later (for example, military recruitment records).13 In order to examine trends in stature from earlier periods of history, researchers must turn to human skeletal remains from archaeological contexts as the primary source of evidence.14 Many studies of past stature have utilized human skeletal remains, including the analyses of hominid remains, temporal trends within and between different regions, the effects of changing subsistence practice, urbanization, climate change, and the impact of disease on human populations.15
A number of authors have undertaken large-scale studies of stature in Roman-period skeletons as a means of assessing physical well-being in different regions and over the duration of the Empire. There is some debate between those who argue for favorable economic and living conditions and those who maintain that the regime was not conducive to good health. Jongman broadly falls into the former camp, asserting that mean femur length was at its longest between the mid-first to mid-second centuries whereas it declined, along with the Empire’s fortunes, during the third and fourth centuries.16 Kron likewise argues for a “healthy” Roman period, stating that male adult stature in Italy during the millennium from 500 BCE to 500 CE was a robust 168cm, which compared favorably with the average stature (163 cm) of nineteenth-century Italian males (historical data from military conscripts).17 However, as Scheidel highlights, Kron’s data-set spans an entire millennium and thus lacks any temporal resolution and nuance.18 Gowland and Garnsey further criticized the regression formulae used in Kron’s study and the potential incompatibility of these with Italian archaeological populations.19 This critique is supported by Gianecchini and Moggi-Cecchi’s study, which suggests that, based on current data, 164cm is a reasonable approximation of mean adult male height in Roman central Italy: similar to that of the nineteenth century conscripts.20 They argue that the formulae frequently used to estimate stature in Roman Italy (Trotter and Gleser’s ‘white’ formulae) result in overestimates, because the Italian archaeological population expresses different body proportions than the skeletal sample used to devise the formulae.21 The lack of reliability in reported stature estimates has led some authors to recommend that stature calculation equations be bypassed altogether, in favor of a direct comparison of long bone length.22 Jantz and Jantz likewise concluded that long bone length is a robust proxy for stature, with lower limb bones tending to show clearer secular trends than upper limb bones.23 Goldewijk and Jacobs analysed a large data-set of c.10,000 burials in their study of stature across the Roman Empire.24 They found that the ratio of femur length to other long bones in their sample was significantly different from those implied in most popular stature reconstruction methods. These authors support the current trend of directly comparing lower limb lengths as a proxy for adult stature.
This chapter reviews the study of stature in the archaeological record with a focus on the Roman period. A critical review of current methods of estimating stature, including an evaluation of the differing techniques and their application to the Roman period will follow. We do not advocate the comparison of long bone lengths alone for the study of stature in the Roman world. We argue that such an approach overlooks the biocultural significance of overall body proportions (including trunk height); instead more effort should be made to establish population-specific regression formulae. Finally, we argue that because adult stature and final body proportions are strongly influenced by environmental conditions during childhood, a more nuanced consideration of growth and adversity during infancy and childhood is required. First, this chapter will provide a detailed discussion of the methods currently used to calculate stature from skeletal remains.
Estimating Stature from Skeletal Remains
The stature of a living individual is a composite of articulated soft and hard tissue elements. Within an archaeological context, human remains are most often skeletonized, frequently incomplete, and unless in situ, are no longer articulated. Since the late nineteenth century, methods for deriving living stature from the “dry bones” of disarticulated skeletons have been produced from documented skeletal collections with known living heights. Pearson created regression formulae to calculate stature from long bone length in 1899 and remarkably this method is still in use by some anthropologists today. Broadly, there are two “types” of method for estimating stature from skeletal remains: the anatomical method (e.g., the “Fully technique”) and the mathematical method.25 A description of each of these, together with their potentials and limitations, is provided below.
THE ANATOMICAL METHOD
The anatomical method was first proposed by Fully in 1956, who used the technique on French World War II prisoners of war for whom ante-mortem records were available.26 This method reconstructs living stature from the measurement of all contiguous skeletal elements that directly contribute to a person’s height, including: cranial height, vertebral height, lengths of the femur and tibia, and the height of the articulated talus and calcaneus of the foot. It additionally incorporates soft tissue corrections to account for the absence of these components in skeletonized remains. Through the measurement of all of those bones that contribute towards height, this technique allows for individual idiosyncrasies in body proportions; for example, short lower limbs compared to long trunk length, or even pathological features impacting on stature, such as spinal curvature, to be taken into consideration.27
In a review of the Fully method, Raxter and colleagues found that it was strongly correlated with living stature, but could underestimate height by as much as 2.4 centimeters, particularly when applied to African American populations.28 These results were thought to relate to inaccuracies in the soft tissue correction factor, which had been derived from European populations. There was also a lack of clarity in terms of the precise anatomical landmarks used by Fully to execute his measurements. Raxter and colleagues have provided useful revisions to Fully’s method, including the creation of new soft tissue corrections.29 In addition, Raxter and colleagues have highlighted the potential source of error introduced through age-related changes in the spinal column and have proposed the use of broad age-correction calculations.30
The primary limiting factor regarding this method is poor preservation and the need for so many contiguous skeletal elements to be present and undamaged. To combat this problem, Auerbach developed methods for estimating values for those skeletal elements that are frequently missing, such as the vertebrae.31 While such estimates can potentially introduce error, these should be minor compared to the much greater error incurred through the use of regression formulae to estimate stature from a single long bone (see below, 2.2).
Overall, the anatomical method provides an effective means of ascertaining population-specific differences in body proportions from well-preserved skeletons. These measurements can then be used to create tailored regression formulae that can be applied to the less complete skeletons. A number of authors have attempted such an approach.32 Unfortunately, due to problems of preservation and the time-consuming nature of this type of analysis, the anatomical method is only infrequently used in archaeological studies.33
THE MATHEMATICAL METHOD
The mathematical method involves the use of simple regression formulae to calculate adult stature from the measurement of single long bones (though the femur and tibia may be used in conjunction if both are present). These formulae are derived from the long bones of documented skeletal individuals for whom living height (or cadaveric stature) is known. Pearson (1899) was the first to produce a “mathematical method” for calculating stature; however, the most widely applied formulae today are those produced by Trotter and Gleser and Trotter (Table 5.1).34 The Trotter and Gleser formulae were created from a sample of white and black Americans from the Terry collection of known individuals in the United States. Formulae were created for the femur, tibia, humerus, radius, and ulna—each with different associated error ranges. Stature estimates calculated from the lower limb bones, in particular the femur, are considered to be the most accurate. Before calculating stature from a skeleton, it is important to first establish the individual’s sex and the (crudely defined) ancestry (i.e., white Americans / black Americans), as the regression equations will differ. In common with the “anatomical method,” an age-related correction factor must also be applied to stature calculated using this method, because individuals (females in particular) display reduced height with age.35
Table 5.1. The regression formulae developed by Trotter and Gleser for estimating stature from the femur
Reference |
Male Formula |
Female Formula |
Trotter and Gleser 1952, 1958 (“White” Formula) |
2.32 * Max Fem Length + 65.53 |
2.47 * Max Fem Length + 54.1 |
Trotter and Gleser 1952, 1958 (“Black” Formula) |
2.10 * Max Fem Length + 72.22 |
2.28 * Max Fem Length + 59.76 |
Trotter 1970 |
2.38 * Max Fem Length + 61.41 |
2.47 * Max Fem Length + 54.1 |
Pearson 1899 |
1.88 * Max Fem Length + 81.306 |
1.95 * Max Fem Length + 72.844 |
Given that this method requires only a single measurement, stature can be calculated significantly more quickly than by the anatomical method and from even very incomplete skeletons.36 However, the accuracy of reconstructed living stature from mathematical regression formulae is affected by intra- and interpopulation differences in body proportions that arise through genetics, ecogeographic variation, and life circumstances.37 The body proportions of the archaeological population may be very different from those of the reference sample on which the formulae were derived. As Table 5.1 illustrates, each of these formulae differ, resulting in stature estimates that can vary by several centimeters. Ideally, one would choose equations based on a reference population that demonstrates similar body proportions, but this is not easily determined or accomplished, as there are only a limited number of skeletal collections available with documented living stature.38 A variety of population-specific regression equations have been produced, including All-brook (British and East African Males), Genovés (modern Mesoamerican and US Southwest), and de Beer (Dutch). However, in practice, the Trotter and Gleser “white” formulae tend to be applied to most archaeological populations, irrespective of either period or place.39
One last point regarding the Trotter and Gleser stature calculation tables involves a methodological controversy. When applying the technique to the original reference population, Jantz noted significant differences between estimated and documented stature.40 Further study concluded that the method described in the original paper was not in fact followed in the creation of the published formulae—of particular concern was the omission of the medial malleolus in measurements of the tibia.41 The implication of this widespread mismeasurement of the tibia for such a prolonged period of time in the forensic and archaeological communities is hard to quantify.42 The following section will consider the applicability of these methods to skeletal samples from late Roman Britain.
APPLYING THE ANATOMICAL AND MATHEMATICAL METHODS TO ADULT SKELETONS: A CASE STUDY FROM ROMAN BRITAIN
Tests of stature calculated via the anatomical method versus the mathematical regression formulae tend to find the former to be the most accurate.43 However, the use of the anatomical method and mathematical method are not mutually exclusive. Vercellotti and colleagues used the anatomical method to create new population-specific regression formulae for European archaeological populations.44 The results were found to be more accurate than using generic regression methods such as Trotter and Gleser’s. We have adopted a similar approach to a skeletal sample from Roman Britain, with the aim of more accurately characterizing body proportions and of arriving at more reliable stature estimations. First, the anatomical method was applied to a total of 76 (36 males and 40 females) well-preserved Romano-British skeletons from five late Roman cemetery sites in southern and eastern England. The resulting calculations formed the basis of “known” stature and body proportions. Secondly, different regression methods were then applied to the sample of 76 individuals and the results compared to the “known” stature (i.e., calculating using the anatomical method). Table 5.2 shows the “known” stature and the deviations from this when each of the commonly used regression formulae were applied. Stature estimations using Trotter and Gleser and Trotter showed a statistically significant difference to the stature calculated using the Fully Anatomical Method (paired t-test; p<0.001) for both males and females.45 When the “black” formula from Trotter and Gleser was used to calculate male stature, the femur provided a significantly different stature (paired t-test; t=–2.1, p<0.01), but there is no significant difference for the tibia (paired t-test; t=1.9, p=0.06).46 This exercise, again, highlights the influence of differential body proportions. When the Pearson formulae were applied, a statistically significant difference was observed for female femoral length (paired t-test; t=2.6, p=0.01), but not for male femoral length (paired t-test; t=–1.1, p=0.28).47 The tibia length equations demonstrated statistically significant differences for both females and males. Finally, when stature was estimated using the sum of the femur and tibia, there was no statistically significant difference with regard to male stature (paired t-test; t=0.5, p=0.64), however a significant difference was observed with the females (paired t-test; t=4.2 p<0.01). This analysis reveals different lower limb body proportions between, not only the Romano-British sample and the Trotter and Gleser and Pearson reference populations, but also males and females.
Finally, population-specific regression formulae were developed from this sample of 76 well-preserved Romano-British based on the “known” body proportions and statures. These regression formulae were then applied to the remaining Romano-British skeletons in the sample which had both a left femur and left tibia preserved. This sample consisted of 174 females and 213 males and mean statures of 154.6 cm (±2.19 cm) for Romano-British females and 164.3 cm (±2.46 cm) for Romano-British males were calculated. These figures are comparable with those estimated by Gianecchini and Moggi-Cecchi for Roman period Central Italy.48
Table 5.2. Stature from Romano-British skeletons calculated using the anatomical method, a range of commonly used regression techniques, and a newly developed population-specific regression method for Roman Britain
The above has demonstrated the utility of the anatomical method for assessing the accuracy of stature estimation techniques and for producing population-specific formulae. However, in this Romano-British sample, the number of skeletons that were sufficiently well-preserved to apply the anatomical method in full was only 35 individuals. The spinal column is an important component of the Fully anatomical method, but taphonomic damage, disease processes, and recovery bias in excavations often lead to missing vertebrae. We were able to more than double this sample to 76 individuals by incorporating estimates for missing vertebral elements. This is possible because the spine exhibits relatively little variation in vertebral body heights between adjacent vertebrae. Individual vertebrae and also whole vertebral sections (e.g., cervical or thoracic sections) can be estimated using mathematical equations based on the size of those vertebrae present.
Vertebral columns with known body heights for all vertebrae (second cervical to fifth lumbar) were used to calculate a coefficient to estimate missing individual vertebral elements. This coefficient was then used to estimate missing vertebrae with the following formula:
where k is the coefficient calculated from known body heights of the vertebra to be estimated, x is the known superior vertebra and y is the inferior vertebra. This calculation allowed the addition of six more individuals, four males and two females, to the “known” sample.
Auerbach produced equations for estimating missing vertebral regions (i.e., cervical, thoracic or lumbar) from known sums of those vertebral sections present.49 A similar approach was trialed on this Romano-British sample, to determine if Auerbach’s equations accurately predicted missing vertebral regions. Vertebral columns with known summed body heights of all vertebrae (C2 through to L5) were entered into Auerbach’s regression formulae for estimating missing regions or total vertebral column length. While the technique generally performed well, there was a statistically significant deviation when estimating the total vertebral column length using Auerbach’s equation.50 New mathematical regression formulae were therefore created to more accurately estimate vertebral column length from the known sum of vertebral regions for this specific Romano-British population. Statistically significant differences were observed between males and females in regard to the summed lumbar vertebral heights. As a result, separate equations were created for males and females.
To estimate the whole vertebral column from the known sum of thoracic and lumbar vertebrae the following equations were created:
1.2216*Sum of Thoracic Sum of Thoracic + 1.0588
*Sum of Lumbar Sum of Lumbar+ 39.333
Males:
1.0801*Sum of Thoracic Sum of Thoracic Sum of Thoracic + 1.3493
*Sum of Lumbar Sum of Lumbar+ 39.921
If thoracic vertebrae were missing, estimation of the whole vertebral column could be estimated from the sum of lumbar vertebrae, though the error associated with this equation is larger than the previous equations.
Females:
2.0395*Sum of LumbarSum of Lumbar + 188.62
Males:
2.8165*Sum of LumbarSum of Lumbar + 98.872
These four equations make possible the addition of 35 individuals to the sample. While the above process may seem rather laborious, it is important to be confident that stature estimates are as accurate as possible and that differential body proportions between populations can be accommodated. The following section will undertake a comparison of long bone lengths only, ignoring the vertebrae entirely.
DIRECT COMPARISON OF LONG BONES
As discussed in the introduction, the direct comparison of long bone lengths, rather than calculated stature data, is now being advocated.51 A direct comparison of average maximum femur length between Romano-British and Anglo-Saxon skeletons shows a significantly reduced femur length for both males and females in Romano-British skeletons when compared to the post-Roman period (Figure 5.1, Table 5.3, t-test: p=0.0001). Furthermore, the mean femur length for males in Roman Britain is very similar to Gianecchini and Moggi-Cecchi’s figure of 445.5mm for males in Roman central Italy.52
One difficulty encountered when using femoral length alone to compare sites is that many skeletal reports simply provide final stature, rather than the raw data (i.e., long bone length). While the current exasperation expressed with the lack of standardization in stature calculation methodology is understandable, and the direct comparison of long bones certainly seems sensible to meet the needs outlined by Goldewijk and Jacobs, we would recommend an approach that considers the full body proportions of a skeletal sample, when possible.53 For example, with the Romano-British and Anglo-Saxon skeletal samples analysed above, we noted differences in mean trunk heights, with the latter having relatively shorter trunk heights (compared to limb lengths) than the former. Therefore, while the Romano-Britons may have had substantially shorter femur lengths, in some instances, this was partially mitigated in terms of overall height by a relatively longer trunk. Goldewijk and Jacobs state that there is “no way of finding out which method renders the correct body heights,” but with the more widespread use of the anatomical method, as described above, it is feasible to characterize the range and variation in body proportions in different regions.54 This analysis of Romano-British stature will be discussed below in the context of recent large-scale studies of Roman-period skeletons.
Table 5.3. Comparison of mean femur length between Romano-British and Anglo-Saxon cemeteries in Southern and Eastern England
Stature in the Roman Empire
Stature as an index of well-being has been widely utilized in studies of Roman-period skeletons. As discussed above, an estimated stature of 168cm for Roman-period males is widely cited and in some instances used to support the suggestion that life at this time incorporated a reasonable standard of living. This estimate is substantially greater than historical data for Italian male conscripts from the nineteenth century—by comparison a measly 163cm. There are a number of criticisms that have been leveled at these estimates, including the choice of regression formulae used to calculate stature. Other criticisms can also be brought to bear; not least a consideration of the actual ages of the Italian conscripts at the time their heights were measured, which was approximately 20 years.55 The adolescent growth spurt is known to be delayed in individuals exposed to adverse circumstances in early life, prolonging growth into the early twenties. Additional height attained during late adolescence is from the trunk, which will continue to grow after the long bones of the limbs have completed fusion. Living conditions for ordinary Italians were far from optimum in the nineteenth century, and the majority of European countries experienced a reduction in mean stature during this period.56 It is therefore very likely that many of these conscripts had not yet finished growing: approximately 9% of final height in males is gained during puberty.57 Some of this putative five-centimeter deficit between ancient Romans and nineteenth-century conscripts will be accounted for by the mismatched comparison of adult (archaeological) with adolescent/young adult (conscripts) data-sets. In addition, and as Gowland and Garnsey have argued, it is highly likely that the incompatibility of the regression formulae used to calculate stature from the Roman-period skeletons is also responsible for an overestimation of male stature.58 A’Hearn and colleagues’ study of growth and stature using the same nineteenth-century military conscript data as Kron further highlights the regional heterogeneity in height at the age of 20 years across Italy.59 Geographically and temporally broad archaeological studies of the Roman Empire should be mindful of subregional complexities in the data and the social and environmental causes of these.
Giannecchini and Moggi-Cecchi suggest that, based on current data, 164cm is a reasonable approximation of mean adult male height in Roman central Italy: similar to that of the nineteenth-century conscripts.60 They argue that the body proportions of the Roman period skeletal samples were more congruent with Trotter and Gleser’s “black” formulae, and furthermore, the use of the “white” formulae produced an overestimate of stature.61 Gianecchini and Moggi-Cecchi’s findings are also remarkably close to the stature calculated independently from the Romano-British sample analysed here using the anatomical method (Table 5.2). Likewise, the Romano-British skeletal sample exhibited body proportions closer to the Trotter and Gleser “black” formulae and were also overestimated by the “white” formulae.
Kopeke studied a sample of Roman skeletons dating from the first to fourth centuries CE and hypothesized that stature would be greatest during the second and third centuries when the Empire was more integrated and stable.62 Her results, however, found that stature was largely unchanged, though it did decline very slightly during the fourth century. Kopeke and Baten argued that stature “stagnates” during the Roman period in central, western, and southern Europe and increases during the fifth and sixth centuries CE.63 Indeed, a number of studies have identified a post-Roman increase in stature across various parts of the Empire.64 For Britain, this tends to be interpreted in terms of an influx of Germanic migrants rather than improved local, environmental conditions. In a similar vein, Koepke and Baten have argued that Roman-period migrants to central Europe (identified as such through grave goods) were on average four cm shorter than the locals.65 This difference in stature between local and nonlocals in both regions may indicate contrasting social and cultural practices, including the greater consumption of milk and meat protein in northern and eastern Europe.66 Interestingly, increased milk consumption is thought to have played a significant part in Japan’s dramatic rise in population stature during the last forty years.67
During the Roman period in Britain the skeletal evidence reveals a decrease in stature compared to the preceding period and an overall increase in nonspecific indicators of health stress.68 A similar skeletal pattern has been observed in Roman-period Italy.69 One contributing factor may be the imposition of an increasingly hierarchical structure onto the local populace with Roman occupation of Britain, resulting in greater social inequalities and increased psycho-social stresses. Scheidel likewise suggests that declining stature in Roman Gaul may be linked to an increase in population size and social inequality.70 Data on stature from modern populations demonstrates a correlation between equality and taller adult stature.71 The stress hormone cortisol is known to inhibit growth and this may be responsible for some of the link between psychosocial stress and reduced stature.72
Adult stature as a measure of well-being lacks nuance because it can mask an array of early life episodes of stress, which have significant implications for morbidity and mortality. A more fruitful approach to the study of Roman health status would be to focus attention on children (i.e., during the growth period). The skeletal remains of children are often overlooked in studies of Roman health.73 Yet children’s bodies are highly sensitive to social and environmental challenges and act as sensitive barometers for overall population health.74 As discussed previously, poor nutrition, infection, and the synergistic interaction between these stressors, are the most influential environmental factors impacting growth.75 Individuals suffering from prolonged episodes of health stress will exhibit a slower rate of bone growth, delayed dental eruption, a prolonged period of growth and ultimately diminished adult stature.76 Growth stunting has been conceptualized in terms of a life-history trade-off in the face of adversities.77 However, in addition to short-term implications, there are significant longer-term health consequences to growth stunting, such as a compromised immune system and impaired cognition.78
Growth, Development and Environment.
The following provides a brief discussion of growth, potential impediments to growth, and the identification of these in the archaeological record.
FETAL AND INFANT GROWTH
At birth the newborn has already had an eventful history that relates most strongly to maternal health and nutritional status.79 Intrauterine conditions have important prolonged implications for growth trajectories in early childhood; therefore it is important to consider the degree to which maternal health can also impact growth and adult stature.80 Women of smaller stature and pelvic dimensions will suffer an increased likelihood of obstetric risks and of giving birth to infants who are ‘small for gestational age’ (SGA).81 Sibley and colleagues’ study of the pelvic dimension of females from a medieval skeletal population in Sudanese Nubia observed a high proportion of females with contracted pelves.82 The authors noted a link between evidence for growth retardation in the mother and neonatal/maternal morbidity and mortality. From a life course perspective, growth stunting of the mother as a consequence of her own poor childhood environment results in the poor health of her offspring, thus perpetuating a cycle of health inequality.83
Studies of growth stunting in developing countries have shown that faltering occurs early in life and is most pronounced by two years of age.84 After birth, there is a period of adjustment in growth as the newborn makes the transition towards a regulatory system based upon its own homeostatic and genetic make-up, rather than the mother’s.85 However, the infant/mother nexus is not yet decoupled; breastfeeding regimes and the nutritional status of the mother continue to exert a strong influence on infant well-being. Poor nutrition profoundly influences growth during fetal, infant, and early childhood development phases—the period of life during which food is predominantly sourced from the maternal body, via the placenta or breast milk.86 A longitudinal study by Chávez and colleagues showed a clear decline in breast milk consumption by the infants of poorly nourished mothers by two to three months of age, contributing to early-childhood malnutrition and poorer growth.87 Height deficits occurring by two years tend to be maintained into adulthood. For example, the four-centimeter secular increase in the adult height of Japanese populations from 1950 to 1990 was already present at the age of two years.88 Thus, Kuzawa and Quinn argue that adult size is most closely linked to matrilineal nutritional well-being and history.89
Postnatal growth velocities are highest during the first year after birth (approximately 30 cm/year in the first two months), but drop dramatically after 12 months.90 Such rapid growth necessitates high dietary requirements during the postnatal period and leaves the infant at greater risk of malnutrition, infection, and death.91 Infant mortality, specifically post-neonatal mortality (death occurring between one month and one year after birth), is strongly correlated with average final adult stature.92 Ideally, therefore, it would be useful to examine infant mortality in conjunction with adult stature as complementary indices of health in the Roman world. While Lewis and Gowland were able to successfully comment on neonatal versus post-neonatal infant mortality in Medieval England, a similar study is not feasible for the Roman period, due to the differential burial treatment of infants.93
The analysis of different skeletal parameters of growth and stress during childhood potentially allows for the production of biographical data. While limb length has been shown to be more ‘plastic’ than trunk height in relation to environmental stresses, differences in vertebral dimensions between archaeological populations have been identified and correlated with increased frailty.94 In other words, growth stunting during infancy, observed in vertebral dimensions, results in increased frailty during adulthood. The transverse and anteroposterior diameters of the neural canal of the vertebrae are “locked-in” by five years of age, due to fusion of this skeletal region, providing an indicator of early post-natal growth, while vertebral body height may continue growing into early adulthood.
The complementary analysis of these parameters, along with long bone length, can be used to construct an osteobiography of growth stunting in early and later childhood, with implications for adult morbidity.95 Intra- and interpopulation comparisons of episodes of growth retardation can be identified and analysed in relation to differing environmental or social variables. Newman and Gowland illustrate the potential of this approach in their study of vertebral dimensions in children from postmedieval London.96 Here, patterns of growth stress were correlated with socially constructed life course norms, including status-driven child-care practices and child labor.
PUBERTY
A mid-childhood growth spurt at around the ages of eight or nine years has been proposed, though this has been contested by some,97 followed by an adolescent growth spurt around puberty. Upon reaching puberty there is an increasing divergence between individuals and sexes (boys experiencing a growth spurt at a later age than girls). Puberty is strongly affected by socioenvironmental conditions, with adverse circumstances leading to delays in growth, pubertal onset, and an extended period of maturation into early adulthood.98 Recently developed osteological techniques now allow age-at-puberty to be assessed from skeletal remains.99 These methods have been applied to studies of Medieval adolescence in England and have important implications for the study of past social age trajectories and fertility. Shapland and colleagues’ study of a large Medieval sample found that the onset of menstruation was delayed until approximately 15 years and over (on the basis of osteological criteria), compared to an average of 13 years today.100 Likewise, male puberty was found to be prolonged, with maturation continuing into the early twenties.101 The first independent application of these new skeletal techniques to Roman-period adolescent skeletons from Britain similarly indicated that puberty was delayed compared to modern norms. A later age of menarche in Romano-British females from rural and urban sites suggests that they were unlikely to have been able to reproduce until their late teens, whilst males were still developing into their early twenties.102 These skeletal data correlated with burial rites accorded some Romano-British females, which indicate a marked change in social status from 18 to 24 years, possibly associated with age-at-marriage, or motherhood.103 The skeletal data is also in line with Galen’s assertions that male development continued until the early to mid-twenties.104 For higher-status groups, with better nutrition, including more dietary protein, age of onset of menarche may well have been much lower, potentially contributing to much younger ages of marriage amongst Roman elites.
CONSTRUCTING GROWTH PROFILES
When analyzing skeletal growth in archaeological populations of children, it is necessary to use dental development age as a proxy for known age. Dental development (not to be confused with dental eruption) is closely correlated with chronological age:105 if a child is unwell for a period of time, or malnourished, his or her teeth will continue to develop, despite stasis in postcranial growth. This observation was established as early as the nineteenth century, when dental age was used to enforce minimum working ages for children in factories.106 Growth parameters such as long bone diaphyseal length can be plotted against dental age to produce growth profiles, thus allowing interpopulation comparisons (Figure 5.2). As always, there are a number of caveats that apply: firstly, dental age, while a close approximation is not a known age and therefore such profiles incorporate any associated error; secondly, these profiles are based on nonsurvivors (i.e., children who died) who may not be representative of the living population. The significance of this latter point has been debated,107 with Saunders and Hoppa arguing that the likely effects are minimal.108 Either way, the comparison of growth profiles between archaeological sites eliminates this as a source of bias because the datasets are comparable.
An example of growth profiles constructed from Romano-British and Anglo-Saxon children is provided in Figure 5.2. Children in both periods fall below the modern values.109 This is true for the majority of growth profiles produced from archaeological populations and is to be expected due to the better nutritional status and reduced infectious disease burden of modern children. The Anglo-Saxon children exhibit growth that is slightly closer to the modern values than the Romano-British sample and this is to be expected given the increased stature in post-Roman Britain. Rohnbogner compared skeletal growth profiles produced from a large sample of Romano-British children with a middle-class eighteenth-to-nineteenth-century population from London.110 Interestingly, she found that the Romano-British children exhibited stronger growth up until approximately five years of age, when the situation reverses. Given the discussion above, this could indicate better maternal health and infant feeding strategies in Roman-Britain compared to industrialized London, but more detrimental conditions in later childhood. Isotopic studies suggest that infants in Roman Britain were generally breastfed until approximately three years of age, with a gradual program of weaning.111 This is much longer than in postmedieval London, and the Roman-British feeding regime may have buffered children against the worst of the environmental circumstances, providing passive immunity, as well as a hygienic and nutritious food source. Interestingly, the isotope evidence shows some differences in infant feeding practices as between Roman Britain and Portus Romae (Isola Sacra), where a shorter period of transitional feeding and earlier cessation of breastfeeding was the norm.112
FIGURE 5.2. Long bone length plotted against dental age for a sample of Romano-British and Anglo-Saxon skeletons. The trend-line for the Romano-British skeletons (dashed) falls below that of the Anglo-Saxons (dotted line).
CORTICAL THICKNESS, VERTEBRAL BODY HEIGHT AND SKELETAL INDICATORS OF ‘STRESS’
In a study of the skeletal remains of children from nineteenth-century Birmingham, England, Mays and colleagues found that the cortical thickness of bones was adversely affected in individuals of lower socioeconomic status.113 Indeed, the authors suggest that appositional growth may be a more sensitive indicator of environmental stressors than longitudinal bone growth. These results were confirmed in a study of postmedieval children by Newman and Gowland, which showed a correlation between poor appositional growth and pathological indicators of poor health, including cribra orbitalia, enamel hypoplasia, rickets, and scurvy (see also Sperduti and colleagues, this volume).114 (Growth patterns and palaeopathological indicators can profitably be considered together in any assessment of the health of individuals.) Figure 5.3 shows a growth profile based on the height of the cervical bodies of the vertebrae. Three individuals exhibit particularly low values relative to their age (skeletons 208, 338, and 262), and it is of note that all three have severe and active pathological indicators of poor health.115
BODY PROPORTIONS
Skeletal growth is heterochronic, meaning that different elements grow at different rates and developmental stages. For example, gains in sitting height (i.e., trunk length) are generally made during infancy and puberty, while gains in leg length occur during childhood. As a consequence, in young children, in respect of growth, legs are ahead of torsos.116 Thus, at one year of age, leg length represents 35% of adult length, but by 10 years this has risen to 77%.117 Therefore poor nutrition or adversity during this period is likely to disproportionately inhibit leg length. A study by Bogin and Baker shows that absolute and relative leg lengths (RLL) provide a summary of postnatal growth between birth and puberty and that RLL is established by eight years of age.118 In general, the timing and duration of a growth insult during childhood can differentially impact upon adult body proportions, resulting in intrapopulation heterogeneity in limb and trunk length.119 A study of living children from highland and lowland Peru is a pertinent illustration of this truth: Pomeroy and colleagues found the ulna and tibia to be most sensitive to poor nutrition.120 Their study would translate well to archaeological data-sets, in the context of studies of the effects of environmental stressors on relative limb segment length, stature and age-at-death in the past. For example, we may hypothesize that the relatively long trunk height identified above in Romano-British individuals compared to the post-Roman period could represent prolonged growth during later adolescence—when most stature increase is derived from the vertebrae. This longer period of growth could be the body’s attempt to mitigate earlier life deficits as a consequence of the higher levels of environmental stressors in Roman compared to post-Roman Britain.
FIGURE 5.3. Vertebral body height for the cervical vertebrae. Dental age in years is plotted on the x-axis and height in millimeters along the y-axis. The Coach Lane and Fewston sites are both postmedieval.
A Holistic Approach to Adult Stature
Adult stature in past societies has tended to be studied cross-sectionally, that is, as a moment in time. However, skeletal growth is diachronic in character, in that particular bones and teeth form at different times and rates. The study of children in the past has the potential to provide insights into how stature was achieved through the analysis of a variety of skeletal parameters relevant to different life course stages. For example, analysis of the body proportions of adults and children, which gives special attention to the proportional lengths of the distal limb segments, may be suggestive of environmental stressors.121 Whilst the effects of growth stunting on individual long bone lengths may be masked by catch-up growth, the analysis of skeletal parameters that fuse earlier in life can provide additional information regarding early childhood.122 Evidence for compensatory growth can then be compiled (e.g., small transverse diameter / distal limb segment / average adult stature) and correlated with mortality risk. This information can be assessed in relation to the presence of childhood indicators of poor health (e.g., cribra orbitalia), with a particular focus on the age of onset of these lesions and the presence of both active and healed lesions within the skeletal sample.123 These data should then be integrated with longitudinal dietary data obtained from incremental isotope analysis of dentine. High resolution isotope analysis of teeth can reveal periods of severe health stress (i.e., elevated nitrogen values) in both the adult survivors and the nonsurvivors who died in childhood.124 Growth deficits in survivors, whose isotope evidence also indicated a period of deprivation, would be of particular interest for correlating with other skeletal parameters of growth stunting. The age at which growth deficits and paleopatholgoical lesions begin to occur is of crucial importance to interpretations. Currently, there is a tendency to interpret such lesions with reference to breastfeeding practices or child care alone. However, if, within a cemetery population the growth of females in the sample was compromised, then growth deficits and paleopathological lesions observed in infants and children could represent an intergenerational legacy, including compromised immune response.125 Social and economic factors such as poverty may carry a heritable biological legacy, resulting in physiological disadvantage that interacts synergistically with social environment to become mutually reinforcing.126
Within the skeletal samples analysed here it is clear that there are not just differences in stature, but also in body proportions between the Romano-British and Anglo-Saxon skeletal samples. It is also apparent that these differences are not only the product of genetics, but higher levels of childhood stress in the earlier period. The timings of these periods of stress in terms of the life course are crucial in terms of those bones that are most affected. Body proportion data can therefore contribute to a biographical understanding of adversity during the period in question. By comparing femur length alone between different cemetery populations, a wealth of additional and pertinent data for examining population well-being is simply being overlooked.
Conclusion
The study of stature as an index of well-being in the Roman world has been beset by problems relating to the lack of standardization in osteological analyses and resulting incompatibility of datasets. Furthermore, there is a tendency within bioarchaeology to adopt a position of parsimonious uniformitarianism, in other words, to apply techniques developed on specific skeletal samples in a universal manner, irrespective of period or place. This approach stems in part from the notion that skeletons are inert biological objects, rather than living tissue, which interacts with social as well as physical environments in a dynamic way.127 Whilst the practicalities and limitations of osteological analysis must be acknowledged, much more can be done to individualize techniques and to strive to construct more informative osteobiographies. This can only be achieved through an approach that incorporates multiple techniques and integrates them within a life course perspective. A life course approach explicitly acknowledges the cumulative and inter-related nature of individual biographies. Therefore, it is recommended that the anatomical method be used when possible to estimate stature from human skeletal remains and to create population-specific formulae. In addition, a variety of skeletal indices should be calculated to establish the difference in limb segment proportions as well as a comparison of trunk and limb lengths. Differences between sexes within the same population may suggest gendered exposure or response to stress, while differences between populations could point to specific biocultural factors. Integration of these data with palaeopathological information and, where possible, isotopic evidence should aid interpretations.
As the above discussion has highlighted, adversity affecting a young girl can impact on her obstetric future and the stature of her offspring and grandchildren. We need to consider the inter-related fortunes of individuals and the range of variables that effect these, otherwise quite basic, physiological parameters. The above discussion has also highlighted the role of nutrition, including infant feeding strategies, in the determination of adult stature. The Roman Empire was a time of high mobility. The calculation of stature in respect of cemetery populations is therefore a complex matter, given that individuals within the sample are likely to have had childhood origins which were diverse. The way ahead, in our view, is to give close attention to the skeletal remains of infants and children, which have tended to be marginalized. As to the mobility of the population in general, the aim should be to arrive at as concrete an analysis of its nature and extent as is possible, making full use of isotopic evidence (via the stable isotopes of strontium, lead, and oxygen), and DNA analysis.
We are sympathetic to the motivations behind the current trend of directly comparing long bones lengths between skeletal samples as a proxy for stature. While such studies have value, we argue that they also have the potential to mislead because they fail to fully account for differential body proportions, including trunk height. Implicit in such studies is the view that different body proportions are a confounding factor—a source of error. We advocate an approach that instead harnesses this complexity with the aim of attaining additional insights into a variety of biocultural factors affecting the growth of individuals at different stages of the life course.
Acknowledgements
We are grateful to Walter Scheidel for inviting us to contribute to this volume and to Tim Thompson for comments on an earlier draft. I am indebted to Peter Garnsey for his encouragement, patience, and wisdom. Finally we are grateful to an anonymous reviewer for helpful comments on an earlier draft.
Notes
1. Sayson and Hargens 2008.
2. Steckel et al. 2002.
3. Vercelloti et al. 2011.
4. Bogin 2001; Steckel et al. 2002.
5. Steckel 2009.
6. Bozzoli et al. 2009, 647.
7. Kemkes-Grottenhaler 2005.
8. Tanner 1974.
9. Blane 2006.
10. Engels 1850, 158.
11. Chadwick 1965.
12. Barker and et al. 2011.
13. Steckel 2004; Gustafsson et al. 2007; Cardoso and Gomes 2009.
14. E.g., Steckel 2007; Steckel and Floud 1997; Roberts and Cox 2010.
15. Frayer 1980; Feldesman et al. 1990 (hominid remains); Eveleth and Tanner 1990; Ruff 1994; de Beer 2004; Steckel 2004; Gustafsson et al. 2007.
16. Jongman 2007.
17. Kron 2005.
18. Scheidel 2012.
19. Gowland and Garnsey 2010. See also Giannecchini and Moggi-Cecchi 2008.
20. Giannecchini and Moggi-Cecchi 2008.
21. Trotter and Gleser 952.
22. Goldewijk and Jacobs 2013.
23. Jantz and Jantz 1999.
24. Goldewijk and Jacobs 2013.
25. E.g., Trotter and Gleser 1952.
26. Fully 1956; Raxter et al. 2006.
27. Raxter et al. 2006; Maijanen 2009; Maijanen and Niskanen 2010; Shin et al. 2012.
28. Raxter, et al. 2006.
29. Raxter et al. 2006; Fully 1956.
30. Raxter et al. 2007.
31. Auerbach 2011.
32. E.g., Scuilli et al. 1990; Scuilli and Giesen 1993; Formicola and Franceschi 1996; Sciulli and Hetland 2007; Raxter et al. 2008.
33. De Mendonça 2000; Raxter et al. 2006; Vercellotti et al. 2009; Auerbach 2011.
34. Pearson 1899; Trotter and Gleser 1952, 1958; and Trotter 1970.
35. Raxter et al. 2006.
36. E.g., Trotter and Gleser 1952, 1958.
37. Vercellotti et al. 2009.
38. Feldesman et al. 1990; Konigsberg et al. 1998; Holliday 1999; Raxter et al. 2006; Sciulli and Hetland 2007; Auerbach and Ruff 2010.
39. Allbrook 1961; Genovés 1967; de Beer 2004; and Trotter and Gleser 1952.
40. Jantz 1992.
41. Jantz 1992; Jantz et al. 1994.
42. Gowland and Thompson 2013.
43. Raxter et al. 2006.
44. Vercellotti et al. 2009.
45. Trotter and Gleser 1952, 1958; and Trotter 1970.
46. Trotter and Gleser 1952, 1958.
47. Pearson 1899.
48. Giannecchini and Moggi-Cecchi 2008.
49. Auerbach 2011.
50. Auerbach 2011.
51. Goldewijk and Jakobs 2013.
52. Gianecchini and Moggi-Cecchi’s 2008.
53. Goldewijk and Jacobs 2013.
54. Goldewijk and Jacobs 2013, 5.
55. A’Hearn et al. 2009.
56. Cole 2003.
57. Karlberg 1998.
58. Gowland and Garnsey 2010.
59. A’Hearn, et al. 2009.
60. Giannecchini and Moggi-Cecchi 2008.
61. Trotter and Gleser 1952, 1958.
62. Kopeke 2002.
63. Kopeke and Baten 2005.
64. E.g., Roberts and Cox 2003; Giannecchini and Moggi-Cecchi 2008. See above, Figure 5.1.
65. Koepke and Baten 2005.
66. King 1999; Scheidel 2012.
67. Takahashi 1984.
68. E.g., Roberts and Cox 2003 Gowland and Redfern 2010; Redfern and DeWitte 2011.
69. E.g., Gianecchini and Moggi-Cecchi 2008.
70. Scheidel 2012.
71. Bozzoli et al. 2009.
72. Walsh 2015.
73. Redfern and Gowland 2012.
74. Lewis 2007.
75. King and Ulijaszek 1999, 161; Humphrey 2000.
76. Humphrey 2000.
77. Bogin et al. 2007.
78. Pelletier 2000; McDade 2003, 2012 (immune system); Chávez 2000; Uauy et al. 2011 (cognition).
79. Tanner 1974; Kuzawa and Bragg 2012.
80. Barker et al. 2001; Barker et al. 2002.
81. Kemkes-Grottenhaler 2005.
82. Sibley et al. 1992.
83. Uauy et al. 2011.
84. Kuzawa and Quinn 2009.
85. Johnston 1986.
86. Kuzawa and Quinn 2009.
87. Chávez and colleagues 2000.
88. Cole 2003.
89. Kuzawa and Quinn 2009.
90. Johnston 1986.
91. Saunders and Barrans 1999, 184.
92. Bozzoli et al. 2009.
93. Lewis and Gowland 2007 (medieval); Gowland et al. 2014 (Roman).
94. Limbs: e.g., Wadsworth et al. 2002. Vertebrae: Watts 2013.
95. Newman and Gowland 2015.
96. Newman and Gowland 2015.
97. E.g., Smith and Buschang 2004.
98. Bogin 1999.
99. Shapland and Lewis 2013, 2014.
100. Shapland, et al. forthcoming.
101. Lewis et al. Forthcoming.
102. Arthur et al. 2016.
103. Gowland 2001.
104. Harlow and Laurence 2002.
105. Saunders 2000.
106. Kirby 2013.
107. Humphrey 2000.
108. Saunders and Hoppa 1993.
109. Maresh 1955.
110. Rohnbogner 2015.
111. Fuller et al. 2006; Powell et al. 2014.
112. Prowse et al. 2008; Powell et al. 2014.
113. Mays et al. 2009.
114. Newman and Gowland 2016.
115. Gowland. Forthcoming.
116. Karlberg 1998.
117. Tardieu 2010, 174.
118. Bogin and Baker 2012.
119. Vercellotti et al. 2011.
120. Pomeroy and colleagues 2012.
121. Pomeroy et al. 2012; Chung and Kuzawa 2014.
122. E.g., transverse diameter of the neural arch; Watts 2013; Newman and Gowland 2015.
123. Walker et al. 2009; DeWitte et al. 2014.
124. Beaumont et al. 2013, 2015; Montgomery et al. 2013.
125. Chung and Kuzawa 2014; Gowland 2015.
126. Chávez et al. 2000.
127. Gowland 2006.
References
A’Hearn, B., F. Peracchi, and G. Vecchi. 2009. “Height and the normal distribution: evidence from Italian military data.” Demography 46: 1–25. DOI: 10.1353/dem.0.0049.
Allbrook, D. 1961. “The estimation of stature in British and East African males.” Journal of Forensic Medicine 8: 15–28.
Arthur, N. A., R. L. Gowland, and R. C. Redfern. 2016. “Coming of age in Roman Britain: osteological evidence for pubertal timing.” American Journal of Physical Anthropology 59: 698–713. DOI: 10.1002/ajpa.22929.
Auerbach, B. M. 2011. “Methods for estimating missing human skeletal element osteometric dimensions employed in the revised Fully technique for estimating stature.” American Journal of Physical Anthropology 145: 67–80. DOI: 10.1002/ajpa.21469.
Auerbach, B. M., and C. B. Ruff. 2010. “Stature estimation formulae for indigenous North American populations.” American Journal of Physical Anthropology 141: 190–207. DOI: 10.1002/ajpa.21131.
Barker, D.J.P., et al. 2001. “Size at birth and resilience to effects of poor living conditions in adult life: longitudinal study.” British Medical Journal 323 (7324): 1273–1276. DOI: 10.1136/bmj.323.7324.1273.
Barker, D.J.P., et al. 2002. “Fetal origins of adult disease: strength of effects and biological basis.” International Journal of Epidemiology 31: 1235–1239.
Barker, D.J.P, et al. 2011. “How boys grow determines how long they live.” American Journal of Human Biology 23 (3): 412–416. DOI: 10.1002/ajhb.21165.
Beaumont, J., et al. 2013. “Victims and survivors: identifying survivors of the Great Famine in 19th century London using carbon and nitrogen isotope ratios.” American Journal of Physical Anthropology 150: 87–98. DOI: 10.1002/ajpa.22179.
Beaumont, J., et al. 2015. “Infant mortality and isotopic complexity: new approaches to stress, maternal health, and weaning.” American Journal of Physical Anthropology 157: 441–457. DOI: 10.1002/ajpa.22736.
Béguelin, M. 2011. “Stature estimation in a Central Patagonian prehispanic population: development of new models considering specific body proportions.” International Journal of Osteoarchaeology 21: 150–158. DOI: 10.1002/oa.1117.
Blane, D. 2006. “The life course, the social gradients, and health.” In Social Determinants of Health (second edition), eds. M. Marmot and R. G. Wilkinson. Oxford: Oxford University Press, 54–77. DOI: 10.1093/acprof:oso/9780198565895.003.04.
Bogin, B. 1999. Patterns of Human Growth. Cambridge: Cambridge University Press.
Bogin, B. 2001. The Growth of Humanity. New York: Wiley-Liss
Bogin, B., and J. Baker. 2012. “Low birth weight does not predict the ontogeny of relative leg length of infants and children: an allometric analysis of the NHANES III sample.” American Journal of Physical Anthropology 148: 487–494. DOI: 10.1002/ajpa.22064.
Bogin, B., M. Inês Varela Silva, and L. Rios. 2007. “Life history trade-offs in human growth: adaptation or pathology?” American Journal of Human Biology 19 (5): 631–642. DOI: 10.1002/ajhb.20666.
Bozzoli, C., A. Deaton, and C. Quintana-Domeque. 2009. “Adult height and childhood disease.” Demography 46: 647–669. DOI: 10.1353/dem.0.0079.
Cardoso, H.F.V., and J.E.A. Gomes. 2009. “Trends in adult stature of peoples who inhabited the modern Portuguese territory from the mesolithic to the late 20th century.” International Journal of Osteoarchaeology 19: 711–725. DOI: 10.1002/oa.991.
Chadwick, E. 1965. The Sanitary Conditions of the Labouring Population of Great Britain: Report 1842. Edinburgh: Edinburgh University Press
Chávez, A., C. Martinez, and B. Soberanes. 2000. “The effect of malnutrition on human development: a 24-year study of well-nourished and malnourished children living in a poor Mexican village.” In Nutritional Anthropology: Biocultural Perspectives on Food and Nutrition, eds. A. H. Goodman, D. L. Dufour, and GH. Pelto. Berkeley: University of California Press, 234–252.
Chung, G. C., and C. W. Kuzawa. 2014. “Intergenerational effects of early life nutrition: maternal leg length predicts offspring placental weight and birth weight among women in rural Luzon, Philippines.” American Journal of Human Biology 26 (5): 652–659. DOI: 10.1002/ajhb.22579.
Cole, T. J. 2003. “The secular trend in human physical growth: a biological view.” Economics and Human Biology 1: 161–168. DOI: 10.1016/S1570–677X(02)00033–3.
de Beer, H. 2004. “Observations on the history of the Dutch physical stature from the late-Middle Ages to the present.” Economics and Human Biology 2: 45–55. DOI: 10.1016/j.ehb.2003.11.001.
de Mendonça, M. C. 2000. “Estimation of height from the length of long bones in a Portuguese adult population.” American Journal of Physical Anthropology 112: 39–48. DOI: 10.1002/(SICI)1096–8644(200005)112:1<39::AID-AJPA5>3.0.CO;2-#.
DeWitte, S. et al., 2014. “Differential survival among individuals with active and healed periosteal new bone formation.” International Journal of Palaeopathology 7: 38–44. DOI: 10.1016/j.ijpp.2014.06.001.
Engels, F. 1950. The Condition of the Working-Class in England in 1844. With a preface written in 1892. London: George Allen and Unwin Ltd.
Eveleth, P. B., and J. M. Tanner. 1990. Worldwide Variation in Human Growth. Second edition. Cambridge: Cambridge University Press.
Feldesman, M. R., J. G. Kleckner, and J. K. Lundy. 1990. “Femur/stature ratio and estimates of stature in mid-and late-Pleistocene fossil hominids.” American Journal of Physical Anthropology 83: 359–372. DOI: 10.1002/ajpa.1330830309.
Feldesman, M. R., and L. K. Lundy. 1988. “Stature estimates for some African Plio-Pleistocene fossil hominids.” Journal of Human Evolution 17 (6): 583–596. DOI: 10.1016/0047–2484(88)90086–3.
Formicola, V. 1993. “Stature reconstruction from long bones in ancient population samples: an approach to the problem of its reliability.” American Journal of Physical Anthropology 90: 351–358. DOI: 10.1002/ajpa.1330900309.
Formicola, V., and M. Franceschi. 1996. “Regression equations for estimating stature from long bones of early Holocene European samples.” American Journal of Physical Anthropology 100: 83–88. DOI: 10.1002/(SICI)1096–8644(199605)100:1<83::AID-AJPA8>3.0.CO;2-E.
Frayer, D. W. 1980. “Sexual dimorphism and cultural evolution in the Late Pleistocene and Holocene of Europe.” Journal of Human Evolution 9: 399–415. DOI: 0.1016/0047–2484(80)90050–0.
Fully, G. 1956. “Une nouvelle méthode de détermination de lataille.” Annales de Medecine Legale 35: 266–273.
Fuller, B. T., et al. 2006. “Isotopic evidence for breastfeeding and possible adult dietary differences from late/sub Roman Britain.” American Journal of Physical Anthropology 129: 45–54. DOI: 10.1002/ajpa.20244.
Gaskell, P. 1833. The Manufacturing Population of England. London: Baldwin and Cradock.
Genoves, S. 1967. “Proportionality of the long bones and their relation to stature among Mesoamericans.” American Journal of Physical Anthropology 26: 67–78. DOI: 10.1002/ajpa.1330260109.
Giannecchini, M., and J. Moggi-Cecchi. 2008. “Stature in archaeological samples from Central Italy: method issues and diachronic changes.” American Journal of Physical Anthropology 135: 284–292. DOI: 10.1002/ajpa.20742.
Goldewijk, G. K., and J. Jacobs. 2013. The Relation between Stature and Long Bone Length in the Roman Empire. Groningen: University of Groningen.
Gowland, R. L. 2001. “Playing dead: implications of mortuary evidence for the social construction of childhood in Roman Britain.” In Proceedings of the Tenth Annual Theoretical Roman Archaeology Conference, eds. G. Davies, A. Gardner, and K. Lockyear Oxford: Oxbow, 152–168.
Gowland, R. L. 2006. “Age as an aspect of social identity: the archaeological funerary evidence.” In Social Archaeology of Funerary Remains, eds. R. L. Gowland and C. Knüsel. Oxford: Oxbow, 143–154.
Gowland, R. L. 2015. “Entangled lives: implications of the developmental origins of health and disease (DOHaD) hypothesis for bioarchaeology and the life course.” American Journal of Physical Anthropology 158: 530–40. DOI: 10.1002/ajpa.22820.
Gowland, R. L., A. T. Chamberlain, and R. C. Redfern. 2014. “On the brink of being: reevaluating infant death and infanticide in Roman Britain.” In Infant Health and Death in Roman Italy and Beyond, eds. M. Carroll, and E.-J. Graham. Portsmouth, RI: Journal of Roman Archaeology, 69–88.
Gowland, R. L. and P. Garnsey. 2010. “Skeletal evidence for health, nutritional status and malaria in Rome and the empire.” Roman Diasporas. Archaeological Approaches to Mobility and Diversity in the Roman Empire, ed. H. Eckardt. Portsmouth, RI: Journal of Roman Archaeology, 131–156.
Gowland, R. L., and S. L. Newman. in press. “Children of the revolution: childhood health inequalities and the life course during industrialisation of the 18th to 19th centuries.” In Children and Childhood in the Past, eds. P. Beauchesne and S. Agarwal. Gainesville: University of Florida Press.
Gowland, R. L., and R. C. Redfern. 2010. “Childhood health at the core and periphery of the Roman Empire.” Childhood in the Past 3: 15–42 Gowland, R. L., and T.J.U. Thompson. 2013. Human Identity and Identification. Cambridge: Cambridge University Press.
Gustafsson, A., et al. 2007. “Stature and sexual dimorphism in Sweden, from the 10th to the end of the 20th century.” American Journal of Human Biology 19 (6): 861–870. DOI: 10.1002/ajhb.20657.
Harlow, M., and R. Laurence. 2002. Growing Up and Growing Old in Ancient Rome: A Life Course Approach. London: Routledge.
Holliday, T. W. 1999. “Brachial and crural indices of European Late Upper Paleolithic and Mesolithic humans.” Journal of Human Evolution 36: 549–566. DOI: 10.1006/jhev.1998.0289.
Humphrey, L. T. 2000. “Growth studies of past populations: an overview and an example.” In Human Osteology in Archaeology and Forensic Science, eds. M. Cox and S. Mays. London: Greenwich Medical Media, 23–38.
Jantz, R. L. 1992. “Modification of the Trotter and Gleser female stature estimation formulae.” Journal of Forensic Sciences 37: 1230–1235. DOI: 10.1520/JFS13310J.
Jantz, R. L., D. R. Hunt, and L. Meadows. 1994. “Maximum length of the tibia: how did Trotter measure it?” American Journal of Physical Anthropology 93: 525–528. DOI: 10.1002/ajpa.1330930410.
Jantz, L. M., and R. L. Jantz. 1999. “Secular change in long bone length and proportion in the United States, 1800–1970.” American Journal of Physical Anthropology 111: 57–67. DOI: 10.1002/(SICI)1096–8644(199909)110:1<57::AID-AJPA5>3.0.CO;2–1.
Johnston, F. E. 1986. “Somatic growth of the infant and preschool child.” In Human Growth, eds. F. Falkner, and J. M. Tanner. New York: Plenum Press, 3–24.
Jongman, W. 2007. “Gibbon was right: the decline and fall of the Roman economy.” In Crises and the Roman Empire, eds. O. Hekster, G. de Kleijn, and D. Slootjes. Brill: Leiden, 183–199. DOI: 10.1163/ej.9789004160507.i-448.38.
Jongman, W. M. 2009. “Archaeology, demography and Roman economic growth.” In Quantifying the Roman Economy: Problems and Methods, eds. A. Bowman, and A. Wilson. Oxford: Oxford University Press, 115–26. DOI: 10.1093/acprof:oso/9780199562596.003.0004.
Karlberg, J. 1998. “The human growth curve.” In The Cambridge Encyclopaedia of Human Growth and Development, eds. S. Ulijaszek, F. Johnston, and M. Preece. Cambridge: Cambridge University Press, 108–113.
Kemkes-Grottenthaler, A. 2005. “The short die young: the inter-relationship between stature and longevity—evidence from skeletal remains.” American Journal of Physical Anthropology 128: 340–347. DOI: 10.1002/ajpa.20146.
King, A. 1999. “Meat diet in the Roman world: a regional inter-site comparison of the mammal bones.” Journal of Roman Archaeology 12: 168–202. DOI: 10.1017/S1047759400017979.
King, S. E., and S. J. Ulijaszek. 1999. “Invisible insults during growth and development.” In Human Growth in the Past, eds. R. D. Hoppa and C. M. Fitzgerald. Cambridge: Cambridge University Press, 161–182.
Kirby, P. 2013. Child Workers and Industrial Health in Britain, 1780–1850. London: Boydell Press.
Koepke, N. 2002. “Regional differences and temporal development of the quality of nutrition in the Roman provinces of Germania and Raetia from the first century to the fourth century AD.” Proceedings of the XIII International Economic History Association Congress. Buenos Aires.
Koepke, N., and J. Baten. 2005. “The biological standard of living in Europe during the last two millennia.” European Review of Economic History 9: 61–95. DOI: 10.1017/S1361491604001388.
Konigsberg, L. W., et al. 1998. “Stature estimation and calibration: Bayesian and maximum likelihood perspectives in physical anthropology.” Yearbook of Physical Anthropology 41: 65–92. DOI: 10.1002/(SICI)1096–8644(1998)107:27+<65::AID-AJPA4>3.0.CO;2–6.
Kron, G., 2005. “Anthropometry, physical anthropology, and the reconstruction of ancient health, nutrition, and living standards.” Historia 54: 68–83. DOI: 10.2307/4436756.
Kuzawa, C. W., and J. M. Bragg. 2012. “Plasticity in human life history strategy: implications for contemporary human variation and the end of genus Homo.” Current Anthropology 53: 369–382. DOI: 10.1086/667410.
Kuzawa, C. W., and W. A. Quinn. 2009. “Developmental origins of adult function and health: evolutionary hypotheses.” Annual Review of Anthropology 38: 131–147. DOI: 10.1146/annurev-anthro-091908–164350.
Lewis, M. E. 2007. The Bioarchaeology of Children. Cambridge: Cambridge University Press.
Lewis, M. E. Forthcoming. “Work and the adolescent in Medieval England (AD 900–1550): the osteological evidence.” Medieval Archaeology.
Lewis, M. E., and R. L. Gowland. 2007. “Brief and precarious lives: infant mortality in contrasting sites from medieval and post-medieval England (AD850–1859).” American Journal of Physical Anthropology 134: 117–129. DOI: 10.1002/ajpa.20643.
Maijanen, H. 2009. “Testing anatomical methods for stature estimation on individuals from the wm bass donated skeletal collection.” Journal of Forensic Science 54:746–752.
Maijanen, H., and M. Niskanen. 2010. “New regression equations for stature estimation for Medieval Scandinavians.” International Journal of Osteoarchaeology 20: 472–480. DOI: 10.1002/oa.1071.
Maresh, M. M. 1955. “Linear growth of long bones of extremities from infancy through adolescence: continuing studies.” American Journal of Diseases in Children 89: 725–742. DOI: 10.1001/archpedi.1955.02050110865010.
Mays, S., R. Ives, and M. Brickley. 2009. “The effects of socioeconomic status on endochondral and appositional bone growth, and acquisition of cortical bone in children from 19th century Birmingham, England.” American Journal of Physical Anthropology 140: 410–416. DOI: 10.1002/ajpa.21076.
McDade, T. W. 2003. “Life history theory and the immune system steps toward a human ecological immunology.” American Journal of Physical Anthropology 46: 100–125. DOI: 10.1002/ajpa.10398.
McDade, T. W. 2012. “Early environments and the ecology of inflammation.” Proceedings of the National Academy of Sciences of the United States of America 109 (2): 17281–17288. DOI: 10.1073/pnas.1202244109.
Minozzie, S., et al. 2013. “‘The Roman Giant’: overgrowth syndrome in skeletal remains from the Imperial Age.” International Journal of Osteoarchaeology 25: 574–584. DOI: 10.1002/oa.2322.
Molleson, T. 1995. “Rates of ageing in the eighteenth century.” In Grave Reflections: Portraying the Past through Cemetery Studies, eds. S. R. Saunders and A. Herring. Toronto: Canadian Scholars Press, 199–222.
Montgomery, J., et al. 2013. “Strategic and sporadic marine consumption at the onset of the Neolithic: increasing temporal resolution in the isotope evidence.” Antiquity 87: 1060–1072. DOI: 10.1017/S0003598X00049863.
Newman, S. L., and R. L. Gowland. 2015. “The use of non-adult vertebral dimensions as indicators of growth disruption and non-specific health stress in skeletal populations.” American Journal of Physical Anthropology 158: 155–164. DOI: 10.1002/ajpa.22770.
Newman, S. L., and R. L. Gowland. Forthcoming. “Dedicated followers of fashion? Bioarchaeological perspectives on socio-economic status, inequality, and health in urban children from the Industrial Revolution.” International Journal of Osteoarchaeology.
Pearson, K. 1899. “Mathematical contribution to the theory of evolution: on the reconstruction of the stature of prehistoric races.” Philosophical Transactions of the Royal Society London 192: 169–244. DOI: 10.1098/rsta.1899.0004.
Pelletier, D. L. 2000. “The potentiating effects of malnutrition on child mortality: epidemiologic evidence and policy implications.” In Nutritional Anthropology: Biocultural Perspectives on Food and Nutrition, eds. A. H. Goodman, D. L. Dufour, and G. H. Pelto. Berkeley: University of California, 227–234.
Pomeroy, E., et al. 2012. “Trade-offs in relative limb length among Peruvian children: extending the Thrifty Phenotypes hypothesis to limb proportions.” PLoS ONE 7 (12): e51795. DOI: 10.1371/journal.pone.0051795
Powell, L. A., et al. 2014. “Infant feeding practices in Roman London: evidence from isotope analysis.” In Infant Health and Death in Roman Italy and Beyond, eds. M. Carroll and E.-J. Graham. Portsmouth, RI: Journal of Roman Archaeology, 89–110.
Prowse, T. L., et al. 2008. “Isotopic and dental evidence for infant and young child feeding practices in an imperial Roman skeletal sample.” American Journal of Physical Anthropology 137: 294–308. DOI: 10.1002/ajpa.20870.
Raxter, M.H., B.M. Auerbach, and C.B. Ruff. 2006. “Revision of the Fully technique for estimating stature.” American Journal of Physical Anthropology 130: 374–384. DOI: 10.1002/ajpa.20361.
Raxter, M. H., C. B. Ruff, and B. M. Auerbach. 2007. “Technical note: Revised Fully stature estimation technique.” American Journal of Physical Anthropology 133: 817–818. DOI: 10.1002/ajpa.20588.
Raxter, M. H., et al. 2008. “Stature estimation in Ancient Egyptians: a new technique based on anatomical reconstruction of stature.” American Journal of Physical Anthropology 136: 147–155. DOI: 10.1002/ajpa.20790.
Redfern, R. C., and S. N. Dewitte. 2011. “Status and health in Roman Dorset: the effect of status on risk of mortality in post-conquest populations.” American Journal of Physical Anthropology 146: 197–208. DOI: 10.1002/ajpa.21563.
Redfern, R. C., and R. L. Gowland. 2012. “A bioarchaeological perspective on the pre-adult stages of the life course: implications for the care and health of children in the Roman Empire.” In Families in the Roman and Late Antique World, eds. M. Harlow and L. L. Loven. New York: Continuum, 111–140.
Roberts, C., and M. Cox. 2007. “The impact of economic intensification and social complexity on human health in Britain from 6000BP (Neolithic) and the introduction of farming to the mid-nineteenth century AD.” In Ancient Health: Skeletal Indicators of Agricultural and Economic Intensification, eds. M. N. Cohen and G.M.M. Crank-Kramer. Gainesville: University of Florida Press, 149–163.
Roberts, C.A., and M. Cox. 2003. Health and Disease in Britain: From Prehistory to the Present Day. Gloucester: Sutton.
Rohnbogner, A. 2015. “Dying young: a palaeopathological analysis of child health in Roman Britain.” PhD Dissertation, University of Reading, England.
Roseboom, T. J., J.H.P. van der Meulen, and A.C.J. Ravelli. 2001. “Effects of prenatal exposure to the Dutch famine on adult disease in later life: an overview.” Molecular and Cellular Endocrinology 185: 93–98.
Ross, M. G., and M. H. Beall. 2008. “Adult sequelae of intrauterine growth restriction.” Seminars in Perinatology 32: 213–218. DOI: 10.1053/j.semperi.2007.11.005.
Ruff, C. B. 1994. “Morphological adaptation to climate in modern and fossil hominids.” American Journal of Physical Anthropology 37: 65–107. DOI: 10.1002/ajpa.1330370605.
Saunders, S. R. 2000. “Subadult skeletons and growth related studies.” In Skeletal Biology of Past Peoples: Research Methods, eds. M. A. Katzenberg and S. R. Saunders. New York: Wiley-Liss, 135–161.
Saunders, S. R., and L. Barrans. 1999. “What can be done about the infant category in skeletal samples?” In Human Growth in the Past, eds. R. D. Hoppa and C. M. Fitzgerald. Cambridge: Cambridge University Press, 183–209.
Saunders, S. R. and R. D. Hoppa. 1993. “Growth deficit in survivors and non-survivors: biological correlates of mortality bias in subadult skeletal samples.” Yearbook of Physical Anthropology 36: 127–151.
Sayson, J. V., and A. R. Hargens. 2008. “Pathophysiology of lower back pain during exposure to microgravity.” Aviation, Space and Environmental Medicine 79: 365–373. DOI: 10.3357/ASEM.1994.2008.
Scheidel, W. 2012. “Physical well-being.” In The Cambridge Companion to the Roman Economy, ed. W. Scheidel. Cambridge: Cambridge University Press, 321–333. DOI: 10.1017/CCO9781139030199.020.
Sciulli, P. W., and M. J. Giesen. 1993. “Brief Communication: an update on stature estimation in Prehistoric Native Americans of Ohio.” American Journal of Physical Anthropology 92: 395–399. DOI: 10.1002/ajpa.1330920309.
Sciulli, P. W., and B. M. Hetland. 2007. “Stature estimation for Prehistoric Ohio Valley Native American populations based on revisions of the Fully Technique.” Archaeology of Eastern North America 35: 105–113.
Sciulli, P. W., K. N. Schneider, and M. C. Mahaney. 1990. “Stature estimation in Prehistoric Native Americans of Ohio.” American Journal of Physical Anthropology 83: 275–280. DOI: 10.1002/ajpa.1330830302.
Shapland F., M. E. Lewis. 2013. “Brief Communication: a proposed osteological method for the estimation of pubertal stage in human skeletal remains.” American Journal of Physical Anthropology 151: 302–310.
Shapland, F., and M. E. Lewis. 2014. “Brief Communication: a proposed method for the assessment of pubertal stage in human skeletal remains using cervical vertebrae maturation.” American Journal of Physical Anthropology 153: 144–153. DOI: 10.1002/ajpa.22416.
Shapland, F., M. Lewis, and R. Watts. Forthcoming. “The lives and deaths of young medieval women: the osteological evidence.” Medieval Archaeology.
Shin, D. H, et al. 2012. “Ancient-to-modern secular changes in Korean stature.” American Journal of Physical Anthropology 147: 433–442.
Sibley, L. M., G. J. Armelagos, and D. P. Van Gerven. 1992. “Obstetric dimensions of the true pelvic in a medieval population from Sudanese Nubia.” American Journal of Physical Anthropology 60: 279–317. DOI: 10.1002/ajpa.1330890403.
Sletner, L., et al. 2014. “Maternal life course socio-economic position and offspring body composition at birth in a multi-ethnic population.” Paediatric and Perinatal Epidemiology 28: 445–454. DOI: 10.1111/ppe.12137.
Smith, S. L., and P. H. Buschang. 2004. “Variation in longitudinal diaphyseal long bone growth in children three to ten years of age.” American Journal of Human Biology 16 (6): 648–657. DOI: 10.1002/ajhb.20077.
Steckel, R. H. 2004. “New light on the ‘Dark Ages’: the remarkably tall stature of Northern European Men during the Medieval Era.” Social Science History 28: 211–228. DOI: 10.1215/01455532–28–2–211.
Steckel, R. H. 2005. “Health and nutrition in the Preindustrial Era: insights from a millennium of average heights in northern Europe.” In Living Standards in the Past, eds. R.C. Allen, T. Bengtsson, and M. Dribe. Oxford: Oxford University Press, 227–254. DOI: 10.1093/0199280681.003.0010.
Steckel, R. H. 2007. A Pernicious Side of Capitalism: The Care and Feeding Slave Children. Available at http://web.econ.ohio-state.edu/rsteckel/vita.pdf.
Steckel, R. H. 2009. “Heights and human welfare: recent developments and new directions.” Explorations in Economic History 46: 1–23. DOI: 10.3386/w14536.
Steckel, R. H., and R. Floud. 1997. Health and Welfare during Industrialization. Chicago: University of Chicago Press.
Steckel, R. H., et al. 2002. “Skeletal health in the Western Hemisphere from 4000 B.C. to the present.” Evolutionary Anthropology 11: 142–155. DOI: 10.1002/evan.10030.
Takahashi, E. 1984. “Secular trend in milk consumption and growth in Japan.” Human Biology 56 (3): 427–437.
Tardieu, C. 2010. “Development of the human hind limb and its importance for the evolution of bipedalism.” Evolutionary Anthropology 19: 174–186. DOI: 10.1002/evan.20276.
Tanner, J. M. 1974. “Variation in growth and maturity of newborns.” In The Effect of the Infant on its Caregiver, eds. M. Lewis and L. A. Rosenblum. New York; London: Wiley-Interscience, 77–103.
Trotter, M. 1970. “Estimation of stature from intact long limb bones.” In Personal Identification in Mass Disasters, ed. T. D. Stewart. Washington, DC: Smithsonian Press, 71–83.
Trotter, M., and G. C. Gleser. 1952. “Estimation of stature from long bones of American whites and Afroamericans.” American Journal of Physical Anthropology 10: 463–512.
Trotter, M., and G. C. Gleser. 1958. “A re-evaluation of estimation of stature based on measurements of stature taken during life and of long bones after death.” American Journal of Physical Anthropology 16: 79–123. DOI: 10.1002/ajpa.1330160106.
Uauy, R., J. Kain, and C. Corvalan. 2011. “How can the developmental origin of health and disease (DOHaD) hypothesis contribute to improving health in developing countries.” American Journal of Clinical Nutrition 96: 17595–17645. DOI: 10.3945/ajcn.110.000562.
Vercellotti, G., et al. 2009. “Stature estimation in an early Medieval (XI–XII c.) Polish population: testing the accuracy of regression equations in a bioarchaeological sample.” American Journal of Physical Anthropology 140: 135–142. DOI: 10.1002/ajpa.21055.
Vercellotti, G., et al. 2011. “Intrapopulation variation in stature and body proportions: social status and sex differences in an Italian medieval population (Trino Vercellese, VC).” American Journal of Physical Anthropology 145: 203–214. DOI: 10.1002/ajpa.21486.
Wadsworth, M. E., et al. 2002. “Leg and trunk length at 43 years in relation to childhood health, diet and family circumstances: evidence from the 1946 national birth cohort.” International Journal of Epidemiology 31: 383–390. DOI: 10.1093/ije/31.2.383.
Walker, P. L., et al. 2009. “The causes of porotic hyperostosis and cribra orbitalia: a reappraisal of the iron deficiencyanemia hypothesis.” American Journal of Physical Anthropology 139: 109–125.
Walsh, J. 2015. “Normal bone physiology, remodeling and its hormonal regulation.” Surgery 33: 1–6. DOI: 10.1016/j.mpsur.2014.10.010.
Watts, R. 2011. “Non-specific indicators of stress and their relationship to age-at-death in medieval York: using stature and vertebral canal neural size to examine the effects of stress occurring during different stages of development.” International Journal of Osteoarchaeology 21: 568–576. DOI: 10.1002/oa.1158.
Watts, R. 2013. “Childhood development and adult longevity in an archaeological population from Barton-upon-Humber, Lincolnshire, England.” International Journal of Palaeopathology 3: 95–104. DOI: 10.1016/j.ijpp.2013.05.001.
Werdelin, L. 1985. “The stature of some medieval Swedish populations.” Fornvännen 80: 133–141.
